Audit the D&D modules

README.md

@@ -31,4 +31,5 @@ Useful references:
 - [Internal implementation docs](docs/internal/overview.md)
 - [Maintained example config](examples/dnd-spells.config.yml)
 - [NPC-grounded example config](examples/dnd-npc-grounded.config.yml)
+- [NPC interaction example config](examples/dnd-npc-interactions.config.yml)
 - [Maintained example input](examples/seriatim-minimal-transcript.json)

docs/config.md

@@ -24,6 +24,8 @@ The explicit-path option is defined in the [CLI reference](cli.md).
 - [D&D NPC configuration](../examples/dnd-npcs.config.yml)
 - [D&D combat-turn configuration](../examples/dnd-combat-turns.config.yml)
 - [D&D NPC-grounded spell and combat configuration](../examples/dnd-npc-grounded.config.yml)
+- [D&D NPC interaction configuration](../examples/dnd-npc-interactions.config.yml)
+- [D&D scene chunk-map configuration](../examples/dnd-scene-chunk-map.config.yml)
 
 All are complete version 3 files. The fragments below illustrate individual
 fields and are not alternate complete configurations.
@@ -330,6 +332,25 @@ During resolution, each selected module's registered option validator runs.
 Production input, chunk, and output bindings reject unknown or invalid options
 with the affected binding context.
 
+### JSON Output Options
+
+The `json` output module accepts only `include_chunk_map`, a boolean that
+defaults to `false`. When `true`, it adds the accepted pipeline-wide chunk map
+to the logical output bundle when one exists. It does not create a CLI flag or
+change output placement.
+
+```yaml
+output:
+  module: json
+  options:
+    include_chunk_map: true
+```
+
+The payload and index descriptor are defined by the
+[Accepted Chunk Map contract](integrations/chunk-map.md). See the complete
+[D&D scene chunk-map configuration](../examples/dnd-scene-chunk-map.config.yml)
+for a copyable pipeline.
+
 Validator bindings use the same shorthand or object module-binding form, but
 only these fields are supported:
 
@@ -355,12 +376,16 @@ production validators do not call the LLM and must not set `llm_profile`.
 | extract | `dnd/spells` | Extracts typed D&D spell-list artifacts. |
 | extract | `dnd/npcs` | Extracts typed D&D NPC-list artifacts. |
 | extract | `dnd/combat-turns` | Extracts typed D&D combat-turn-list artifacts. |
+| extract | `dnd/npc-interactions` | Extracts typed D&D NPC-interaction-list artifacts from a supplied NPC registry. |
+| extract | `dnd/scene-descriptions` | Extracts one typed D&D scene description for each accepted chunk. |
 | merge | `appendorder` | Combines typed artifacts in chunk order. |
 | normalize | `noop` | Passes merged typed artifacts through unchanged. |
 | normalize | `dnd/spells` | Deterministically canonicalizes and de-duplicates typed D&D spell-list artifacts. |
 | normalize | `dnd/npcs` | Deterministically consolidates typed D&D NPC-list artifacts by canonical name and unions exact evidence. |
 | normalize | `dnd/combat-turns` | Deterministically canonicalizes, orders, and de-duplicates typed D&D combat-turn artifacts. |
-| output | `json` | Produces JSON output files for normalized `application/json` lanes. |
+| normalize | `dnd/npc-interactions` | Canonicalizes registry NPC names, orders interaction occurrences, and removes only exact duplicates. |
+| normalize | `dnd/scene-descriptions` | Validates, orders, trims, and de-duplicates typed D&D scene-description artifacts. |
+| output | `json` | Produces JSON output files for normalized `application/json` lanes and can opt in to an accepted chunk map. |
 
 ## Implemented Production Validators
 
@@ -382,6 +407,15 @@ production validators do not call the LLM and must not set `llm_profile`.
 | `extract/dnd/combat-turns/source_refs` | deterministic | Rejects missing or invalid D&D combat-turn source references. |
 | `extract/dnd/combat-turns/source_relatedness` | deterministic | Emits warnings when an actor is not found near cited source text. |
 | `normalize/dnd/combat-turns/invariants` | deterministic | Rejects normalized combat-turn identity, evidence-order, and chronology violations. |
+| `extract/dnd/npc-interactions/shape` | deterministic | Rejects malformed D&D NPC-interaction-list artifacts. |
+| `extract/dnd/npc-interactions/registry` | deterministic | Rejects interaction names absent from the supplied NPC registry. |
+| `extract/dnd/npc-interactions/source_refs` | deterministic | Rejects missing, invalid, or extract-chunk-external D&D interaction source references. |
+| `extract/dnd/npc-interactions/source_relatedness` | deterministic | Emits bounded warnings when an interaction name is not found in its cited source text. |
+| `normalize/dnd/npc-interactions/invariants` | deterministic | Rejects normalized interaction identity, evidence-order, and chronology violations. |
+| `extract/dnd/scene-descriptions/shape` | deterministic | Rejects malformed D&D scene-description-list artifacts. |
+| `extract/dnd/scene-descriptions/source_refs` | deterministic | Rejects invalid current-source references and extract records not exactly attached to their chunk. |
+| `extract/dnd/scene-descriptions/source_relatedness` | deterministic | Emits bounded advisory warnings for title or summary text not grounded in its cited transcript range. |
+| `normalize/dnd/scene-descriptions/invariants` | deterministic | Rejects non-normalized scene order, exact duplicates, and ID or range conflicts. |
 
 The production default chain for `dnd/spells` is used for both its extract and
 normalize stages:
@@ -436,6 +470,48 @@ normalize:
     - extract/dnd/combat-turns/source_relatedness
 ```
 
+The production default chains for `dnd/npc-interactions` are:
+
+```yaml
+extract:
+  validators:
+    - generic/valid_json
+    - extract/dnd/npc-interactions/shape
+    - extract/dnd/npc-interactions/registry
+    - extract/dnd/npc-interactions/source_refs
+    - generic/valid_json_schema
+    - extract/dnd/npc-interactions/source_relatedness
+normalize:
+  validators:
+    - generic/valid_json
+    - extract/dnd/npc-interactions/shape
+    - extract/dnd/npc-interactions/registry
+    - normalize/dnd/npc-interactions/invariants
+    - extract/dnd/npc-interactions/source_refs
+    - generic/valid_json_schema
+    - extract/dnd/npc-interactions/source_relatedness
+```
+
+The production default chains for `dnd/scene-descriptions` are:
+
+```yaml
+extract:
+  validators:
+    - generic/valid_json
+    - extract/dnd/scene-descriptions/shape
+    - extract/dnd/scene-descriptions/source_refs
+    - generic/valid_json_schema
+    - extract/dnd/scene-descriptions/source_relatedness
+normalize:
+  validators:
+    - generic/valid_json
+    - extract/dnd/scene-descriptions/shape
+    - normalize/dnd/scene-descriptions/invariants
+    - extract/dnd/scene-descriptions/source_refs
+    - generic/valid_json_schema
+    - extract/dnd/scene-descriptions/source_relatedness
+```
+
 Empty chains approve output by default.
 
 The `generic` chunker accepts:
@@ -492,6 +568,13 @@ accepts no references. The maintained
 accepted normalized output to the later spell and combat targets through an
 explicit ordered step.
 
+The `dnd/scene-descriptions` extractor accepts only the optional `players`,
+`party`, and `glossary` campaign slots. They can disambiguate terms in the
+model-facing title and summary but are never source evidence. Its normalizer
+accepts no references. No NPC registry or generated artifact is required. See
+the [D&D scene-description artifact contract](integrations/dnd-scene-description-artifacts.md)
+and the copyable [scene-description configuration](../examples/dnd-scene-descriptions.config.yml).
+
 The `dnd/combat-turns` extractor declares the optional campaign slots and the
 structured `npcs` slot. Campaign references guide only the LLM extraction
 stage. The deterministic normalizer declares only `npcs`, whose operation-time
@@ -516,6 +599,16 @@ identity; names, content, and paths are not recorded there. When
 absent, the combat prompt receives the exact empty registry value
 `{"npcs":[]}` with its projection digest and no registry provenance.
 
+Both `dnd/npc-interactions` stages require the structured `npcs` slot. Bind one
+accepted normalized NPC artifact from an earlier ordered step; the interaction
+extractor receives only its names-only projection, while the normalizer uses
+the immutable registry for exact canonical-name lookup. A missing, rejected,
+or incompatible producer prevents the interaction step from running. Registry
+source references remain identity provenance and never supply interaction
+evidence: every interaction must cite current transcript units. The complete
+durable contract and two-step configuration are defined in the
+[D&D NPC interaction artifact contract](integrations/dnd-npc-interaction-artifacts.md).
+
 ## State Surfaces
 
 The `output`, `cache`, and `debug` top-level fields select independent physical

docs/integrations/chunk-map.md

@@ -0,0 +1,83 @@
+# Accepted Chunk Map
+
+This document defines the durable accepted chunk-map artifact that the JSON
+output encoder can write as `chunk-map.json`. It describes the exact accepted,
+materialized chunks used by a run; it is not a lane artifact and is never an
+input to later pipeline steps. Enable it with the JSON output option described
+in [Configuration](../config.md#json-output-options).
+
+## Identity
+
+- Artifact kind: `source/chunk-map`
+- Logical file: `chunk-map.json`
+- Schema ID: `notarius.source.chunk_map`
+- Schema name: `notarius_source_chunk_map_v1`
+- Schema version: `v1`
+- Media type: `application/json`
+
+The checked-in [JSON Schema](../../internal/framework/chunkmap/assets/schemas/source_chunk_map.v1.json)
+defines the strict wire shape. `chunk-map.json` is listed by the optional
+`chunk_map` descriptor in [the JSON output index](json-output.md#indexjson),
+not by the lane-oriented `output_files` collection.
+
+## Payload
+
+The payload has these required fields:
+
+- `source_id`: accepted source-document identity.
+- `source_digest`: canonical lower-case `sha256:` digest of that document.
+- `plan_digest`: canonical lower-case `sha256:` digest of the accepted logical
+  plan.
+- `requested_chunker`: chunk module selected by the current resolved pipeline.
+- `producer`: the original accepted-plan producer, with required
+  `input_module` and `chunk_module`; `llm_profile` is present only for an
+  LLM-backed producer.
+- `plan_annotations`: accepted plan-level annotation namespace map. It is
+  `{}` when no namespaces are present.
+- `chunks`: non-empty execution-order collection of accepted chunks.
+
+Each chunk has `id`, zero-based `index`, `source_ref`, positive `unit_count`,
+and an explicit `annotations` namespace map. A source reference has the source
+ID and inclusive positive `start_unit_id` and `end_unit_id` endpoints.
+Annotation values are arbitrary valid JSON under non-empty namespaces. They
+are preserved as canonical JSON without interpreting any module-specific
+namespace.
+
+## Invariants
+
+The framework constructs this artifact only after materializing the selected
+logical plan and accepting it through the configured chunk validator chain.
+Construction proves the source and plan digests, source-document range order,
+materialized chunk IDs and indexes, source references, unit membership and
+counts, and plan/range annotations agree exactly. Chunk IDs are unique and
+indexes are contiguous and agree with array order.
+
+The codec rejects unknown fixed-object fields, malformed identities or
+digests, invalid annotation JSON, trailing JSON content, and any payload whose
+reconstructed logical plan does not reproduce `plan_digest`. It makes
+defensive copies at serialization and decoding boundaries.
+
+## Acceptance And Provenance
+
+The artifact is available only when the chunk plan was accepted. It remains
+available when a later extraction, merge, or normalization result is rejected;
+it is absent when chunk validation rejects the candidate plan.
+
+`requested_chunker` describes the current pipeline selection. `producer`
+describes who originally produced the accepted plan. On a cache hit these can
+differ: the accepted ranges, annotations, digests, and stable materialized IDs
+are reused, while the producer remains the stored producer. Cache paths,
+actions, references, metadata, warnings, timestamps, and detailed provenance
+remain in the run manifest rather than this payload.
+
+## Data Handling
+
+The map contains structure, not source content. It excludes transcript bytes,
+materialized units, source-unit metadata, chunk content, private model
+responses, rejected proposals, debug data, external-reference content, and
+filesystem paths.
+
+Annotations can nevertheless be source- or model-derived. Treat an enabled
+`chunk-map.json` with the same sensitivity and retention expectations as lane
+output. Physical placement, confined atomic writing, and permissions follow
+the ordinary [output operation](../operations.md#output).

docs/integrations/dnd-npc-artifacts.md

@@ -5,6 +5,10 @@ the selectable production NPC pipeline. The normalized JSON payload can be
 passed explicitly to the spell extractor as an optional caster-name registry
 or to the combat extractor and normalizer as an actor registry. It
 remains a reference, not spell or combat evidence.
+The NPC interaction extractor and normalizer also consume this registry for
+canonical identity; registry source references remain provenance and never
+become interaction evidence. Their occurrence contract is defined in the
+[D&D NPC interaction artifact](dnd-npc-interaction-artifacts.md).
 
 ## Identity
 

docs/integrations/dnd-npc-interaction-artifacts.md

@@ -0,0 +1,148 @@
+# D&D NPC Interaction Artifact
+
+This document defines the durable D&D NPC-interaction-list artifact and its
+two-step production pipeline. It records discrete, source-grounded occurrences
+for NPCs already accepted into a normalized NPC registry; it does not expand
+the registry or summarize events.
+
+## Identity And JSON
+
+- Artifact kind: `dnd/npc-interaction-list`
+- Durable schema ID: `notarius.dnd.npc_interactions`
+- Durable schema name: `notarius_dnd_npc_interactions_v1`
+- Durable schema version: `v1`
+- Media type: `application/json`
+
+The payload is one strict JSON object with only an `interactions` array. The
+array may be empty. Each item has exactly `name`, `kind`, and `source_refs`:
+
+```json
+{
+  "interactions": [
+    {
+      "name": "Mira Thorn",
+      "kind": "dialogue",
+      "source_refs": [
+        {"source_id": "session-7", "start_unit_id": 12, "end_unit_id": 13}
+      ]
+    }
+  ]
+}
+```
+
+`name` is the canonical display name from the required NPC registry.
+`source_refs` contains one or more current-source ranges with required
+`source_id`, `start_unit_id`, and `end_unit_id`; unit IDs are positive integers.
+During extraction, every range must be wholly contained in the current accepted
+chunk. This prevents a candidate from citing valid units that were not presented
+to that extraction call.
+Unknown fields are rejected.
+
+## Interaction Categories
+
+`kind` is exactly one of:
+
+| Kind | Meaning |
+| --- | --- |
+| `mentioned` | The NPC is referred to, but is not established as present or communicating in the evidenced passage. |
+| `noncombat_presence` | The NPC is present and relevant to the passage but does not meaningfully participate in dialogue or combat. |
+| `dialogue` | The NPC speaks, responds, or is directly engaged in a meaningful non-combat exchange. |
+| `combat_ally` | The NPC actively participates in combat on the party's side. |
+| `combat_opponent` | The NPC actively participates in combat against the party. |
+| `other` | The transcript clearly establishes a direct NPC occurrence that fits none of the preceding kinds. |
+
+`other` is a residual category for positively evidenced activity, not a fallback
+for uncertain classification. When activities overlap, active combat
+participation outranks dialogue, presence, and mention; dialogue outranks
+non-combat presence and mention; and non-combat presence outranks mention.
+Combat alignment is not resolved by precedence: a meaningful change between
+ally and opponent creates separate occurrences.
+
+These categories do not encode summaries, relationships, state, motives, or
+unobserved events.
+
+## Occurrence Boundaries And Ordering
+
+One occurrence represents one NPC, one kind, and one locally coherent passage
+within one accepted chunk. Repeated evidence belongs to the same occurrence
+only while it supports the same uninterrupted activity. A kind change, combat
+alignment change, intervening scene or meaningful absence, or transition from
+mention to presence starts a new occurrence. Occurrences never span chunks, and
+merge or normalization never semantically combines nearby, overlapping, or
+cross-chunk records.
+
+Normalization orders records by:
+
+1. earliest valid source-document position;
+2. the NPC identity comparison key;
+3. the exact canonical NPC display name;
+4. interaction kind in lexical order; and
+5. the complete canonical source-reference sequence, ordered by source ID and
+   the source-document positions of each range's start and end.
+
+Only records with identical canonical names, kinds, and complete valid evidence
+sequences are duplicates. Different categories, ranges, or separately grounded
+occurrences remain separate.
+
+## Evidence, Registry, And Normalization
+
+The registry proves only the canonical NPC identity. Its source references are
+registry provenance and are never interaction evidence. Every durable
+interaction must cite current transcript units supporting both the name and its
+classification.
+
+The extractor receives a names-only registry projection such as
+`{"npcs":[{"name":"Mira Thorn"}]}`. The normalizer uses the full immutable
+registry for exact canonical-name lookup. It canonicalizes source references
+and applies the ordering and exact-duplicate rules above.
+
+## Production Pipeline
+
+The extractor and normalizer key is `dnd/npc-interactions`. Both require the
+structured `npcs` slot, so an accepted normalized registry must come from an
+earlier step:
+
+```yaml
+pipelines:
+  dnd-npc-interactions:
+    input: seriatim
+    steps:
+      - id: identify-npcs
+        artifacts:
+          npcs:
+            extract: dnd/npcs
+            normalize: dnd/npcs
+      - id: extract-interactions
+        references:
+          npcs:
+            artifact:
+              step: identify-npcs
+              lane: npcs
+        artifacts:
+          interactions:
+            extract: dnd/npc-interactions
+            normalize: dnd/npc-interactions
+```
+
+The framework passes only the accepted normalized producer. A missing, rejected,
+or incompatible NPC artifact prevents the consumer step from executing. It
+records generated artifact identity and bounded producer provenance without
+copying registry names, source ranges, or payload content into the manifest.
+
+## Validation And Metadata
+
+The default extract chain is `generic/valid_json`, interaction shape, registry,
+and source-reference validation, `generic/valid_json_schema`, then warning-only
+source relatedness. The normalize chain runs normalized invariants after
+registry validation and before source-reference and schema validation, followed
+by relatedness. Normalizer and relatedness warnings are bounded and end with an
+omission summary when necessary. The codec metadata contains only
+`interaction_count`. Extractor metadata identifies its prompt and private
+response schema; component-local checkpoint identities include the names-only
+registry projection where relevant. Generated registry identity stays in
+framework provenance and dependency fingerprints.
+
+See [Configuration](../config.md#implemented-production-modules) for selectable
+keys and chains, [the NPC artifact contract](dnd-npc-artifacts.md) for the
+registry boundary, and the copyable
+[maintained example](../../examples/dnd-npc-interactions.config.yml).

docs/integrations/dnd-scene-description-artifacts.md

@@ -0,0 +1,120 @@
+# D&D Scene Description Artifact Contract
+
+This document defines the durable `dnd/scene-description-list` artifact
+emitted by the D&D scene-description lane.
+
+## Artifact identity
+
+| Property | Value |
+| --- | --- |
+| Artifact kind | `dnd/scene-description-list` |
+| Schema ID | `notarius.dnd.scene_descriptions` |
+| Schema name | `notarius_dnd_scene_descriptions_v1` |
+| Schema version | `v1` |
+| Media type | `application/json` |
+
+The normalized payload is written by the JSON output module to
+`lanes/<lane-id>.json`. See [JSON output](json-output.md) for the surrounding
+output bundle.
+
+## JSON shape
+
+The payload is a JSON object containing exactly one required field, `scenes`.
+Each scene object contains exactly these required fields:
+
+| Field | Shape and ownership |
+| --- | --- |
+| `id` | Non-empty accepted chunk ID, assigned by Notarius. |
+| `source_ref` | Exact inclusive accepted chunk range, assigned by Notarius. |
+| `kind` | One of `combat`, `narrative`, `recap`, or `meta`. |
+| `title` | Non-empty, trimmed, source-grounded title. |
+| `summary` | Non-empty, trimmed, source-grounded summary. |
+
+All object layers reject unknown fields. The `scenes` array must be present and
+non-empty. `source_ref` has exactly `source_id`, `start_unit_id`, and
+`end_unit_id`; its source ID is non-empty and its unit IDs are positive
+integers.
+
+```json
+{
+  "scenes": [
+    {
+      "id": "chunk-000001",
+      "source_ref": {
+        "source_id": "session-alpha",
+        "start_unit_id": 1,
+        "end_unit_id": 3
+      },
+      "kind": "narrative",
+      "title": "Arrival at the watchtower",
+      "summary": "The party reaches the ruined watchtower and begins to investigate it."
+    }
+  ]
+}
+```
+
+`id` and `source_ref` are application-owned identity and evidence. The model
+provides only `kind`, `title`, and `summary`; it is not asked for chunk IDs,
+source IDs, unit IDs, ranges, participants, or confidence.
+
+## Scene kinds
+
+| Kind | Meaning |
+| --- | --- |
+| `combat` | Active combat is a substantive central activity. |
+| `narrative` | Current-session in-world play that is not principally combat, recap, or meta discussion. |
+| `recap` | The table is primarily recounting a previous session. |
+| `meta` | Sustained out-of-character discussion is the scene's primary purpose. |
+
+For a mixed accepted chunk, classification prefers `combat`, then `recap`,
+then `meta`, then `narrative`. Brief table talk, rules clarification, or a
+short recollection does not replace the main current-session activity.
+
+## Extraction and evidence
+
+The extractor runs once for each accepted chunk and maps one successful model
+response to one scene record. It copies the current chunk ID and exact chunk
+range, preserves the model kind without repair, and trims only surrounding
+whitespace from title and summary. A model response cannot represent an empty
+result; extraction failure follows the configured retry and rejection policy.
+
+Optional `players`, `party`, and `glossary` campaign references can help
+disambiguate names or setting terms. They never supply scene evidence or add
+events absent from the accepted chunk. The lane requires no NPC registry or
+other generated artifact reference.
+
+## Merge and normalization
+
+Accepted per-chunk lists are appended in chunk order. Normalization then:
+
+1. validates the current-source range, non-empty ID and prose, and closed kind;
+2. trims only title and summary whitespace;
+3. sorts records by source-document start position, then ID;
+4. removes records only when all five durable fields are identical;
+5. rejects a reused ID when any remaining durable field differs; and
+6. rejects the same exact range when `kind`, `title`, or `summary` differs.
+
+Two different IDs with the same range and identical model-owned content remain
+separate records. Normalization does not join adjacent ranges, rewrite prose,
+repair kinds, infer missing scenes, or use chunk annotations.
+
+## Validation and warnings
+
+Extraction validation requires exactly one record with an ID and range exactly
+equal to its current chunk. Later validation checks source membership without a
+current chunk. Shape, source-range, kind, ID, and normalized-invariant failures
+reject the artifact.
+
+Relatedness checks are advisory. They separately warn when a scene title or
+summary has no significant lexical token in its cited transcript range. The
+check ignores short tokens and common function words, uses transcript text only,
+and does not treat campaign references as evidence. Warning diagnostics are
+bounded and do not copy transcript or campaign-reference content.
+
+## Configuration
+
+Use `dnd/scenes` to form accepted scene chunks, then select
+`dnd/scene-descriptions` for extraction and normalization. The maintained
+[scene-description configuration](../../examples/dnd-scene-descriptions.config.yml)
+is a copyable minimal lane. Selectable keys and default validator chains are
+defined in [Configuration](../config.md).

docs/integrations/json-output.md

@@ -18,6 +18,8 @@ The encoder writes:
 - `lanes/<lane-id>.json`, one file per normalized serialized artifact
 - `rejected.json`
 - `warnings.json`
+- `chunk-map.json`, only when the JSON output binding enables
+  `include_chunk_map` and the run has an accepted chunk map
 
 Files are pretty-printed JSON with a trailing newline when the payload is JSON.
 Logical file paths are relative, slash-separated, and may not contain `..`.
@@ -60,6 +62,13 @@ contains the normalized payload `media_type`, normalizer `module_key`, and
 response `schema_id`, `schema_name`, and `schema_version` when those values are
 available.
 
+When present, the top-level optional `chunk_map` descriptor contains exactly
+`artifact_kind`, `file`, `media_type`, `schema_id`, `schema_name`, and
+`schema_version`. It identifies the pipeline-wide `chunk-map.json`; it is not
+a lane output and never appears in `output_files`. The descriptor and file are
+both absent when export is disabled or no chunk plan was accepted. Its payload
+contract is defined by [Accepted Chunk Map](chunk-map.md).
+
 ## `manifest.json`
 
 `manifest.json` contains a run manifest. This abridged example shows its core
@@ -153,7 +162,8 @@ contains the codec-owned JSON bytes pretty-printed.
 The schema of each lane payload is owned by that artifact contract. For the
 current D&D lanes, see [D&D Spell Artifact](dnd-spell-artifacts.md),
 [D&D NPC Artifact](dnd-npc-artifacts.md), and
-[D&D Combat-Turn Artifact](dnd-combat-turn-artifacts.md).
+[D&D Combat-Turn Artifact](dnd-combat-turn-artifacts.md), and
+[D&D Scene Description Artifact](dnd-scene-description-artifacts.md).
 
 ## `rejected.json`
 

docs/internal/llm.md

@@ -39,7 +39,8 @@ without exposing Scriptorium types through stage contracts.
 7. injecting that one shared client into complete pipeline preparation before
    the source file is read or the runner is invoked.
 
-The D&D scene chunker and spell, NPC, and combat extractors retain this
+The D&D scene chunker and spell, NPC, combat-turn, NPC-interaction, and
+scene-description extractors retain this
 injected client and use it for every structured completion. Operation requests
 do not carry an LLM client.
 
@@ -98,7 +99,8 @@ D&D prompt content. `internal/framework/promptfs` provides the domain-neutral
 filesystem composition helper used to combine module-owned files with shared
 domain prompt fragments.
 
-The D&D scene chunker and spell, NPC, and combat-turn extractors each declare an
+The D&D scene chunker and spell, NPC, combat-turn, NPC-interaction, and
+scene-description extractors each declare an
 ordered prompt asset manifest. The manifest lists the package-owned YAML and
 Markdown files, then the exact shared fragments rendered by that prompt; the
 same ordered list drives both filesystem mounting and the prompt fingerprint.
@@ -126,21 +128,31 @@ per-chunk transcript changes. Cache reuse requires the preceding message
 sequence and content to be exactly identical; semantic similarity is not
 sufficient. Cache boundaries belong at the ends of reusable stable tiers,
 subject to the provider's cache-boundary limit. The shared identity and
-campaign-reference messages form the first two extraction boundaries. Spell
-and combat prompts add a boundary at the shared NPC registry. Each extraction
+campaign-reference messages form the first two extraction boundaries. Spell,
+combat, and interaction prompts add a boundary at the shared NPC registry. Each extraction
 prompt places its final boundary on its lane-specific instructions, immediately
 before the transcript. The transcript does not carry cache control because no
 reusable content follows it.
 
-Accordingly, the common prefix of all three extraction prompts is system,
+Accordingly, the common prefix of the spell, NPC, combat, and interaction
+extraction prompts is system,
 extraction evidence, identity, and campaign references. The NPC prompt then
 renders task, instructions, and transcript. Spell renders the NPC registry,
 catalog, task, instructions, and transcript. Combat renders the NPC registry,
-task, instructions, and transcript. The
+task, instructions, and transcript. NPC interaction renders the names-only NPC
+registry, task, instructions, and transcript. The
 scene chunker is not an extraction lane: it retains its separate system,
 transcript, campaign-reference, task, and instruction order and marks its
 transcript and campaign-reference messages ephemeral.
 
+The scene-description extractor deliberately omits the citation-oriented
+`common-dnd-extraction-evidence.md` asset because Notarius attaches the whole
+accepted chunk range itself. Its manifest is system, shared identity, shared
+campaign references, lane task, lane instructions, then the transcript. The
+identity, campaign-reference, and instruction messages are ephemeral cache
+boundaries; the transcript is last and has no cache control. Compatible shared
+messages remain canonical shared assets rather than copied package text.
+
 Shared wording belongs in the canonical assets under
 `internal/modules/dnd/shared`; extraction packages reference those assets in
 their manifests instead of copying similar text into package-local files.
@@ -154,7 +166,8 @@ return defensive copies, and expose a diagnostics map that omits schema bytes.
 The small framework registry contains only generic test schemas; production
 schemas remain package-owned.
 
-The spell, NPC, and combat extractors' package-owned prompts declare their
+The spell, NPC, combat, NPC-interaction, and scene-description extractors'
+package-owned prompts declare their
 structured JSON inputs and private response schemas. Each private response
 schema remains separate from its durable artifact codec schema; this work does
 not use shared schema fragments or schema generation. Those private schemas own
@@ -167,14 +180,14 @@ numbers. The spell extractor's prompt declares a required
 the catalog input from its prepared
 effective catalog as `{"spell_names":[...]}` using sorted canonical names only.
 The shared D&D prompt assets include a generic NPC grounding fragment directly
-after the campaign reference message for spell and combat prompts. When an NPC
+after the campaign reference message for spell, combat, and interaction prompts. When an NPC
 registry is bound, the
 domain registry boundary strictly decodes and identity-validates one durable
 artifact, re-encodes canonical JSON for provenance, and separately generates a
 names-only prompt projection. The unbound projection is exactly `{"npcs":[]}`.
 Prompt input and component-local checkpoint digests cover the projected bytes;
 manifests retain the optional full registry digest/count rather than names,
-overlay bytes, registry paths, or source metadata. Combat prompt,
+overlay bytes, registry paths, or source metadata. Combat and interaction prompt,
 response-schema, mapping, normalization, identity, and registry-projection
 fingerprints remain separate semantic inputs to checkpoint identity.
 

docs/internal/modules.md

@@ -17,16 +17,16 @@ validator registry. Package-family registrars compose those leaf registrations
 into the production catalog and own family-level policy such as default
 validator chains and prompt asset collection.
 
-Production input, chunk, output, and D&D spell- and combat-extract packages
+Production input, chunk, output, and D&D spell-, NPC-, combat-, interaction-, and scene-description-extract packages
 register strict option decoders and run-local builders. Preparation decodes their options into
 implementation-owned values and injects dependencies plus the materialized
 reference set for the selected target. Each builder receives an isolated clone
-of that set; input and output builders receive no references. The spell and
-combat extractors are typed over the canonical D&D model. D&D validators, merge,
+of that set; input and output builders receive no references. The spell, NPC,
+combat, interaction, and scene-description extractors are typed over the canonical D&D model. D&D validators, merge,
 and normalize use typed variants; JSON representation validators use serialized
 requests; and unconditional validators expose separate chunk and typed
 variants. The D&D production registrar registers the canonical typed spell,
-NPC, and combat implementations, including their kind-specific merge and
+NPC, combat, interaction, and scene-description implementations, including their kind-specific merge and
 normalize behavior.
 
 For D&D artifact defaults, generic JSON syntax validation runs first. Rejecting
@@ -54,7 +54,7 @@ package-local embedded assets. Shared filesystem composition belongs in
 `internal/framework/promptfs`; reusable D&D prompt fragments, reference
 declarations, prompt-input assembly, and source-unit/citation helpers belong in
 `internal/modules/dnd/shared`, which also owns bounded D&D diagnostics. The
-D&D scene chunker and spell, NPC, and combat-turn extractors use ordered
+D&D scene chunker and spell, NPC, combat-turn, NPC-interaction, and scene-description extractors use ordered
 package-local prompt manifests for both rendering and prompt fingerprinting, so
 only the shared fragments each prompt actually renders participate in either
 operation. Extraction prompts place stable shared and lane-specific context
@@ -65,7 +65,7 @@ contracts expose only Notarius structured-completion types, not Scriptorium
 public types.
 
 The shared `ChunkPromptMaterial` helper owns common transcript material
-preparation for the spell, NPC, and combat-turn extractors. It clones supplied
+preparation for the spell, NPC, combat-turn, and NPC-interaction extractors. It clones supplied
 source metadata, falls back to the materialized chunk when content is absent,
 checks that content remains chunk-identical, and fills only the common default
 fields. Extractors retain their request validation and wrap helper errors with
@@ -122,6 +122,18 @@ The `internal/modules/dnd/codec/combatturns` package owns the durable
 is registered by the production D&D family registrar for the selectable combat
 lane.
 
+The `internal/modules/dnd/codec/npcinteractions` package owns the durable
+`dnd/npc-interaction-list` schema and candidate versus approved JSON boundary.
+It is registered by the production D&D family registrar for the selectable
+interaction lane. Its external contract is documented in the
+[D&D NPC interaction artifact contract](../integrations/dnd-npc-interaction-artifacts.md).
+
+The `internal/modules/dnd/codec/scenedescriptions` package owns the durable
+`dnd/scene-description-list` schema and candidate versus approved JSON boundary.
+It is registered by the production D&D family registrar. Its external contract
+is documented in the
+[D&D scene-description artifact contract](../integrations/dnd-scene-description-artifacts.md).
+
 ## Input Adapter
 
 ### `internal/modules/seriatim/input/transcript`
@@ -165,17 +177,16 @@ window settings used by `Plan`.
 
 The scene chunker prepares a structured Scriptorium request from the full
 transcript, session, and optional D&D reference inputs. It validates the model's
-scene boundaries against source-unit IDs and converts them into deterministic
-plan ranges with optional scene annotations. Preparation injects the shared
-structured LLM client into the chunker; `Plan`
-supplies only the run-specific profile, session, source, references, and
-metadata.
+inclusive source-unit endpoints against document position and converts them
+into deterministic plan ranges. Preparation injects the shared structured LLM
+client into the chunker; `Plan` supplies only the run-specific profile, session,
+source, references, and metadata.
 
 Scene validation requires sequential, contiguous, non-overlapping coverage from
-the first source unit through the last. Scene descriptions, boundaries,
-confidence, and participants are module-owned annotations. Boundary caveats
-become warnings. Malformed
-structured output is returned as an error; there is no fallback chunker.
+the first source unit through the last. Its private response contains only the
+boundary endpoints; the accepted plan has no D&D-specific annotations and
+produces no boundary warnings. Malformed structured output is returned as an
+error; there is no fallback chunker.
 
 The package embeds its prompt and response schema and reports their non-secret
 identity and hashes through singleton module metadata. Shared D&D assets supply
@@ -262,6 +273,18 @@ integer source-unit candidates unchanged. Required semantic content, non-empty
 evidence, and valid source ranges are rejected by the deterministic shape and
 source-reference validators.
 
+### `internal/modules/dnd/extract/scenedescriptions`
+
+The scene-description extractor makes one structured completion for each
+accepted chunk and maps its private `kind`, `title`, and `summary` response to
+one `dnd.SceneDescription`. It assigns the current chunk ID and exact range,
+preserves kind without repair, and trims only title and summary whitespace.
+Optional players, party, and glossary references can disambiguate prompt terms
+but do not supply evidence. The package owns its private schema, prompt assets,
+and mapping fingerprint; deterministic validators own the durable semantic
+checks. The durable contract is defined in the
+[D&D scene-description artifact contract](../integrations/dnd-scene-description-artifacts.md).
+
 ### `internal/modules/dnd/extract/combatturns`
 
 The combat extractor prepares one structured request per supplied chunk using
@@ -284,6 +307,28 @@ The combat normalizer accepts only the optional structured NPC registry.
 Campaign references remain extractor-only LLM context and are not materialized
 for deterministic normalization.
 
+### `internal/modules/dnd/extract/npcinteractions`
+
+The NPC interaction extractor requires the structured `npcs` registry slot. It
+uses the registry's names-only prompt projection with shared extraction
+evidence, identity, and transcript material, then maps private model records to
+`dnd.NPCInteractionList` with the current source identity. Registry source
+references are never reused as interaction evidence. The private response
+schema carries only name, bounded interaction kind, and source-unit ranges;
+deterministic validators own registry membership, source validity, and
+relatedness. Extract-stage source validation additionally requires every cited
+range to be wholly contained in the current materialized chunk. Prompt, schema,
+mapping, and the names-only registry projection
+participate in checkpoint identity, while generated producer identity remains
+framework provenance.
+
+The domain-owned `internal/modules/dnd/npcinteractions` package defines
+canonical source-reference and occurrence ordering, valid-evidence eligibility,
+and collision-safe exact identity. The interaction normalizer and normalized
+invariants validator both consume those rules, so their production and checking
+paths cannot drift. Normalizer and relatedness warning lists use the shared D&D
+diagnostic cap and emit a final omission-summary warning when truncated.
+
 ### `internal/modules/dnd/normalize/npcs`
 
 The NPC normalizer performs deterministic identity-aware consolidation in
@@ -344,6 +389,25 @@ digest/count metadata; generated identity is retained in framework provenance
 and dependency fingerprints. The normalizer is included in the production D&D
 registrar with the default combat normalization chain.
 
+### `internal/modules/dnd/normalize/npcinteractions`
+
+The interaction normalizer requires the same immutable NPC registry. It
+canonicalizes exact registry-name matches, orders and de-duplicates source
+references, stable-sorts occurrences by source-document position, and collapses
+only exact interaction identities with valid evidence. It does not infer,
+merge, or summarize distinct occurrences. Its metadata and fingerprints expose
+the normalization and NPC identity policies; generated registry identity stays
+in framework provenance and checkpoint dependencies.
+
+### `internal/modules/dnd/normalize/scenedescriptions`
+
+The scene-description normalizer has no options or references. It validates
+each source range against the source document, trims title and summary
+whitespace, orders records by source position then ID, removes only exactly
+identical records, and rejects conflicting reused IDs or ranges. Its policy
+fingerprint identifies this deterministic behavior; the matching invariant
+validator checks the normalized result in the production chain.
+
 ## Output Encoder
 
 ### `internal/modules/generic/output/json`
@@ -353,6 +417,12 @@ safe logical names, pretty-prints JSON payloads, and assembles the logical index
 manifest, rejected-result, warning, and lane files. Invalid JSON, unsupported
 media types, unsafe names, and sanitized-name collisions are errors.
 
+Its strict `include_chunk_map` option is disabled by default. When enabled, it
+validates the framework-supplied accepted chunk map through its codec and adds
+the pipeline-wide `chunk-map.json` plus its index descriptor; it does not treat
+the map as a lane payload. The external shape is owned by the
+[Accepted Chunk Map contract](../integrations/chunk-map.md).
+
 The encoder returns logical files only. The CLI places them on disk, and the
 [JSON output contract](../integrations/json-output.md) defines their external
 paths and schemas.
@@ -428,6 +498,30 @@ expose local policy fingerprints. In the registered defaults, JSON syntax runs
 first; combat shape, normalized invariants when applicable, and source-reference
 validation precede JSON Schema validation; warning-only relatedness runs last.
 
+## D&D NPC Interaction Validators
+
+Interaction shape validation owns the required list, registry name, supported
+kind, and non-empty source-reference collection. Registry validation checks
+exact membership in the required immutable NPC registry. Source-reference and
+relatedness validation use the current transcript only; malformed candidates
+are deferred by later validators and produce no relatedness warning. The
+normalized-invariants validator owns canonical registry names, source-reference
+order, chronology, and exact duplicate identity. The production chains run
+shape, registry, and source-reference checks before JSON Schema validation;
+relatedness remains warning-only and last.
+
+## D&D Scene Description Validators
+
+Scene-description shape validation owns the non-empty list, trimmed ID and
+prose, closed kind, and basic source-reference shape. Extract-stage source
+validation additionally requires the one record to attach exactly to the
+current accepted chunk; later source validation checks source membership.
+Relatedness checks the title and summary independently against only their cited
+transcript range and emits bounded advisory warnings. The normalized-invariants
+validator owns ordering, exact duplicate elimination, and conflicting ID or
+range detection. The production chains run shape and source-reference checks
+before JSON Schema validation; the warning-only relatedness check is last.
+
 ## Production Registration
 
 Production composition occurs through family registrars. The CLI allocates one
@@ -436,7 +530,7 @@ complete framework registry set and one LLM asset registry. It invokes
 `internal/modules/seriatim/register`, and `internal/modules/dnd/register` in
 that order, then exposes the matching catalog for resolution. The generic and
 Seriatim registrars own their production leaf registrations. The D&D registrar
-owns D&D leaf registrations, typed spell, NPC, and combat default-validator
+owns D&D leaf registrations, typed spell, NPC, combat, interaction, and scene-description default-validator
 chains, typed append-order specializations, and D&D prompt/schema asset
 collection. Its registration helpers group module, validator, prompt-asset, and
 chain composition while retaining artifact-specific merge and clone behavior in

docs/internal/overview.md

@@ -50,6 +50,7 @@ required normalized artifacts have crossed the typed handoff.
 | `internal/framework/promptfs` | Builds module prompt filesystems from module-owned and caller-provided shared prompt assets. |
 | `internal/framework/checkpoint` | Root-based checkpoint loading, recording, identity, and payload serialization. |
 | `internal/framework/chunkplan` | Source-addressed chunk-plan filesystem storage, envelope validation, and atomic publication. |
+| `internal/framework/chunkmap` | Strict durable accepted chunk-map construction, schema, validation, cloning, and serialization. |
 | `internal/framework/debug` | Root-based framework and LLM debug recording. |
 
 Framework contracts provide typed artifact, provenance-wrapper, chunk-validator,
@@ -67,8 +68,10 @@ run-local construction closures. Preparation injects shared dependencies and
 constructs input, chunk, validators, ordered lanes, and output before source
 parsing. Production modules use strict construction-time option decoding, and
 LLM-backed modules retain the injected shared client. The D&D family registers
-the canonical `dnd/spell-list`, `dnd/npc-list`, and `dnd/combat-turn-list`
-codecs, typed spell, NPC, and combat extractors and normalizers, validators,
+the canonical `dnd/spell-list`, `dnd/npc-list`, `dnd/combat-turn-list`,
+`dnd/npc-interaction-list`, and `dnd/scene-description-list` codecs, typed
+spell, NPC, combat, interaction, and scene-description extractors and
+normalizers, validators,
 plus kind-specific generic merge strategies; generic JSON validators use the
 serialized-validation contract. The runner executes lanes through
 private exact-type-checked closures, coordinates extract results independently
@@ -87,15 +90,24 @@ Configuration. The implemented module packages are:
 | `internal/modules/seriatim/input/transcript` | Parses the supported Seriatim transcript format into the generic source model. |
 | `internal/modules/generic/chunk/units` | Splits ordered source units by unit count and overlap. |
 | `internal/modules/dnd/chunk/scenes` | Produces contiguous D&D scene chunks from structured model output. |
-| `internal/modules/dnd` | Owns the canonical D&D spell-list, spell-cast, NPC-list, NPC, combat-turn-list, and combat-turn artifact types. |
+| `internal/modules/dnd` | Owns the canonical D&D spell-list, spell-cast, NPC-list, NPC, combat-turn-list, combat-turn, NPC-interaction-list, and scene-description-list artifact types. |
 | `internal/modules/dnd/codec/spells` | Strictly decodes and stably encodes the durable D&D spell-list representation. |
 | `internal/modules/dnd/codec/npcs` | Strictly decodes and stably encodes the durable D&D NPC-list representation. |
 | `internal/modules/dnd/codec/combatturns` | Strictly decodes and stably encodes the durable D&D combat-turn-list representation. |
+| `internal/modules/dnd/codec/npcinteractions` | Strictly decodes and stably encodes the durable D&D NPC-interaction-list representation. |
+| `internal/modules/dnd/codec/scenedescriptions` | Strictly decodes and stably encodes the durable D&D scene-description-list representation. |
 | `internal/modules/dnd/extract/spells` | Maps private structured model output to canonical source-grounded D&D spell lists. |
 | `internal/modules/dnd/extract/npcs` | Maps private structured model output to canonical source-grounded D&D NPC lists. |
 | `internal/modules/dnd/extract/combatturns` | Maps private structured model output to source-grounded D&D combat-turn candidates and preserves chronology and invalid candidate values for validators. |
+| `internal/modules/dnd/extract/npcinteractions` | Maps private structured model output to current-source NPC interaction candidates grounded by a required registry. |
+| `internal/modules/dnd/extract/scenedescriptions` | Maps one private scene description to the current accepted chunk's ID and exact range. |
+| `internal/modules/dnd/npcinteractions` | Owns canonical source-reference ordering, occurrence ordering, valid-evidence checks, and exact interaction identity shared by normalization and invariant validation. |
 | `internal/modules/dnd/normalize/combatturns` | Canonicalizes and orders merged combat turns, applies exact NPC identity matches, and collapses only exact valid-evidence duplicates. |
+| `internal/modules/dnd/normalize/npcinteractions` | Canonicalizes required-registry names, orders interaction occurrences, and collapses only exact valid-evidence duplicates. |
+| `internal/modules/dnd/normalize/scenedescriptions` | Trims, source-orders, and removes only exactly identical scene descriptions while rejecting ID and range conflicts. |
 | `internal/modules/dnd/validate/combatturns` | Provides deterministic shape, source-reference, source-relatedness, and normalized-invariant validation for the production combat chains. |
+| `internal/modules/dnd/validate/npcinteractions` | Provides deterministic shape, registry, source-reference, source-relatedness, and normalized-invariant validation for interaction chains. |
+| `internal/modules/dnd/validate/scenedescriptions` | Provides deterministic shape, exact extraction attachment, source-relatedness, and normalized-invariant validation for scene-description chains. |
 | `internal/modules/dnd/npcs/registry` | Resolves validated normalized NPC references into immutable grounding data and exact identity lookup. |
 | `internal/modules/dnd/npcs/identity` | Owns Unicode-aware NPC identity, ID derivation, and registry collision validation. |
 | `internal/modules/dnd/spells/catalog` | Embeds and validates the versioned D&D 5e 2014 SRD catalog, composes optional overlays, and provides immutable effective lookup. |
@@ -103,7 +115,7 @@ Configuration. The implemented module packages are:
 | `internal/modules/generic/normalize/noop` | Preserves accepted merged output. |
 | `internal/modules/dnd/normalize/spells` | Canonicalizes catalog-backed spell names and exact source references, conservatively collapses duplicate casts, and reports deterministic warnings and independently scoped catalog checkpoint identity. |
 | `internal/modules/dnd/normalize/npcs` | Consolidates NPC records deterministically by canonical name, unions exact evidence, and reports bounded warnings. |
-| `internal/modules/generic/output/json` | Encodes manifests, lane payloads, warnings, and rejections as logical JSON files. |
+| `internal/modules/generic/output/json` | Encodes manifests, lane payloads, warnings, rejections, and an explicitly enabled accepted chunk map as logical JSON files. |
 
 `internal/modules/dnd/shared` owns reusable D&D prompt fragments,
 reference declarations, prompt input assembly, source-unit reference helpers,
@@ -116,14 +128,16 @@ The `dnd/npcs/registry` package owns the optional `npcs` registry boundary.
 External references are strictly decoded and identity-validated during
 preparation; generated references are decoded and identity-validated at the
 ordered step handoff. Both paths retain canonical registry JSON for provenance
-and emit a names-only projection to operation-time spell and combat prompts.
-Combat normalization uses the canonical registry for exact actor lookup. The
+and emit a names-only projection to operation-time spell, combat, and
+interaction prompts. Combat and interaction normalization use the canonical
+registry for exact name lookup. The
 framework records generated identity and bounded producer provenance, while
 the raw external reference remains independently tracked by pipeline
 provenance. An absent registry is represented only by the empty prompt value
 `{"npcs":[]}`. Spell
 and combat consumers use this shared boundary without changing their public
-module contracts.
+module contracts. Interaction consumers require it and retain only current
+transcript references as durable evidence.
 
 Generic validators under `internal/modules/generic/validate` provide
 unconditional test decisions, JSON syntax validation, and JSON Schema

docs/internal/pipeline.md

@@ -119,10 +119,10 @@ extractor's artifact kind, so variants under one reusable key may declare
 different capabilities and reference slots. Kind-neutral registry inspection
 selects the first registered artifact kind in sorted order.
 
-Production composition registers the D&D spell-list codec and typed extractor,
-matching typed merge, normalize, and semantic-validator variants, and
-serialized JSON validators. Every artifact lane resolves through the typed
-registries and a matching codec.
+Production composition registers the D&D spell-list, NPC-list, combat-turn-list,
+NPC-interaction-list, and scene-description-list codecs and typed lane
+variants, plus serialized JSON validators. Every artifact lane resolves through
+the typed registries and a matching codec.
 
 A `ModuleSpec` declares its stage plus required and provided capabilities.
 Chunk, extract, merge, and normalize specs may also declare reference slots.
@@ -208,15 +208,18 @@ The runner:
 3. selects a stored plan or executes the configured chunker's `Plan` operation;
 4. canonicalizes and materializes the plan, then validates the resulting
    chunks;
-5. executes each resolved step in configuration order. For one step, it
+5. builds the framework-owned accepted chunk map from the accepted source,
+   logical plan, and exact materialized chunks, then supplies it to the output
+   request independently of output-module options;
+6. executes each resolved step in configuration order. For one step, it
    dispatches extract jobs in source-chunk then resolved-lane order, starts a
    bounded lane continuation when all extracts for that lane are terminal, and
    waits for every lane to become terminal;
-6. encodes and validates each accepted normalized producer artifact, then
+7. encodes and validates each accepted normalized producer artifact, then
    builds the immutable generated reference sets for the next step;
-7. invokes the prepared output encoder only after every step succeeds and
+8. invokes the prepared output encoder only after every step succeeds and
    validates its logical file results;
-8. returns the assembled manifest, outcomes, warnings, and files.
+9. returns the assembled manifest, outcomes, warnings, and files.
 
 Within each artifact lane, it reuses the prepared extractor, merger, normalizer,
 and validators while performing these transitions:
@@ -265,6 +268,14 @@ currently requested chunker and the effective plan producer. Cache state and
 paths are configured and operated outside the runner; see
 [Configuration](../config.md#state-surfaces) and [Operations](../operations.md).
 
+For an accepted plan, the runner also constructs the strict framework-owned
+[Accepted Chunk Map](../integrations/chunk-map.md) before lane execution. It
+uses the current resolved chunker as `requested_chunker` and the stored or
+generated record as `producer`, preserving that distinction on reuse. Chunk
+rejection supplies no map; later lane rejection does not discard it. Output
+encoders receive a defensively owned serialized value and may explicitly
+ignore it.
+
 The extract job channel has the same capacity as the effective extract worker
 count, so dispatch applies backpressure. A fixed continuation executor prevents
 ready or checkpoint-reused lanes from creating one goroutine each. Workers and

docs/internal/state.md

@@ -17,6 +17,11 @@ debug roots.
 The pipeline runner returns logical output files. After validating every
 logical name, the CLI exclusively creates the run directory beneath the
 selected output root and performs confined, atomic file writes within it.
+The runner supplies an accepted chunk map as an optional, defensively owned
+output-request artifact. The JSON encoder alone decides whether its explicit
+option writes the map and optional index descriptor; neither the map payload
+nor its annotations are copied into the run manifest. The durable fields are
+owned by the [Accepted Chunk Map contract](../integrations/chunk-map.md).
 
 `internal/framework/chunkplan` owns source-addressed plan storage, validation,
 and atomic publication. Its store is constructed only when the selected mode is

docs/operations.md

@@ -37,6 +37,12 @@ automatically removes output. The
 [JSON output contract](integrations/json-output.md) owns the logical file
 names, schemas, and media types inside a run directory.
 
+An enabled JSON `include_chunk_map` option adds an accepted chunk map to durable
+output. Its annotations may contain source- or model-derived data, so retain
+and protect it like lane output. The map is opt-in and does not alter existing
+bundles; its payload exclusions are defined in the
+[Accepted Chunk Map contract](integrations/chunk-map.md).
+
 Remove an output run directory only after its consumer data is no longer
 needed. This is data deletion, not cache cleanup.
 

docs/roadmap/audit.md

@@ -0,0 +1,766 @@
+# D&D Extraction Module Refactoring Audit Strategy
+
+Status: Strategy complete; final audit results recorded below
+
+## Purpose
+
+Define a disciplined audit of the five production D&D extraction modules:
+
+- `dnd/spells`;
+- `dnd/npcs`;
+- `dnd/combat-turns`;
+- `dnd/npc-interactions`; and
+- `dnd/scene-descriptions`.
+
+The audit will determine whether these modules follow a coherent set of
+conventions, whether repeated implementation can be replaced by appropriately
+scoped shared code or assets, and whether the accumulated design suggests other
+maintainability improvements.
+
+This document defines how to perform the audit. It does not contain audit
+findings and does not authorize production changes.
+
+## Audit Principles
+
+The audit must distinguish consistency from uniformity. The five modules should
+use the same conventions where they perform the same responsibility, but a
+module should remain different when its artifact semantics, evidence model,
+reference requirements, or normalization policy require it.
+
+Recommendations must follow these principles:
+
+- Prefer evidence from current code, tests, prompts, schemas, configuration,
+  and documentation over naming or visual similarity.
+- Treat a divergence as a finding only when it is unexplained, increases
+  maintenance or correctness risk, or violates an intended convention.
+- Do not recommend an abstraction solely to reduce line count. A shared helper
+  must own one coherent invariant and make future correct changes easier.
+- Preserve package ownership of artifact semantics, private response DTOs,
+  private structured-output schemas, and lane-specific prompts.
+- Keep D&D-specific behavior in D&D packages. Move behavior into a generic
+  framework package only when the contract is demonstrably domain-neutral and
+  has a non-D&D consumer or a clear framework-owned responsibility.
+- Evaluate prompt sharing by byte identity and semantic ownership. Prompt
+  caching benefits only when repeated message content and ordering are exactly
+  identical.
+- Apply the testing policy to proposed refactors. Prefer behavioral protection
+  at stable boundaries and do not add change-detector tests for helper usage,
+  prompt length, exact hashes, or private file layout.
+
+## Scope
+
+### Primary scope
+
+Inspect the complete package-owned implementation beneath:
+
+- `internal/modules/dnd/extract/spells`;
+- `internal/modules/dnd/extract/npcs`;
+- `internal/modules/dnd/extract/combatturns`;
+- `internal/modules/dnd/extract/npcinteractions`; and
+- `internal/modules/dnd/extract/scenedescriptions`.
+
+For each package, include:
+
+- module identity, capabilities, construction, options, registration, and
+  execution class;
+- reference-slot declarations and construction-time or operation-time
+  reference handling;
+- checkpoint fingerprints and manifest metadata;
+- request validation and structured LLM request preparation;
+- private response DTOs, response-schema loading, and response mapping;
+- source-reference resolution, evidence attachment, canonicalization,
+  ordering, and exact deduplication;
+- embedded prompt manifests, shared and local assets, message order, cache
+  boundaries, and schema assets;
+- errors, warnings, diagnostics, cloning, and mutation safety; and
+- package-local tests and test support.
+
+### Contextual scope
+
+Inspect a neighboring component only when needed to determine ownership,
+duplication, or compatibility:
+
+- `internal/modules/dnd/shared` and focused D&D subpackages used by more than
+  one extractor;
+- the five artifact model and codec contracts;
+- corresponding merge, normalize, and validate variants;
+- production registration and default validator composition;
+- pipeline reference, fingerprint, and LLM contracts;
+- canonical current-behavior documentation and integration contracts; and
+- representative production and integration tests.
+
+Contextual inspection is not a request to redesign every lane stage. Findings
+outside extraction should be reported only when they directly explain an
+extractor inconsistency or reveal a misplaced responsibility.
+
+### Exclusions
+
+Do not use this audit to:
+
+- change durable artifact schemas or extraction policy;
+- redesign the fixed pipeline shape or ordered-step model;
+- combine distinct artifacts into a larger D&D result;
+- evaluate live-model output quality;
+- introduce schema generation, a dependency-injection framework, or a general
+  module superclass;
+- move domain rules into the generic framework;
+- rewrite tests merely to make their file layout look alike; or
+- implement any recommended refactor.
+
+If the audit exposes a product-contract concern, record it separately from
+refactoring recommendations and identify the additional decision required.
+
+## Comparison Method
+
+### 1. Establish a module inventory
+
+Create one row per module in a working comparison matrix. Record exact current
+facts rather than inferred conventions:
+
+| Dimension | Facts to record |
+| --- | --- |
+| Identity | Module key, artifact kind, capabilities, execution class |
+| Files | Production files, embedded assets, focused test files |
+| Construction | Dependencies, options, reference decoding, immutable prepared state |
+| Registration | `ModuleSpec`, builder, option validation, declared slots |
+| Provenance | Manifest metadata and checkpoint fingerprint keys and values |
+| Prompt | Prompt ID/version, manifest messages, inputs, shared assets, cache boundaries |
+| Schema | Private schema identity, strictness, loader, diagnostics behavior |
+| Execution | Request validation, LLM call, response mapping, errors and warnings |
+| Evidence | Source identity, range resolution, canonicalization, ordering, deduplication |
+| Tests | Contract owner, malformed cases, integration coverage, test-only helpers |
+
+Use the matrix to identify exact agreement, intentional variation, and
+unexplained variation. Do not infer a preferred convention from whichever
+module was implemented first. Determine the preferred shape from architecture,
+current documentation, shared contracts, and the clearest implementation.
+
+### 2. Classify every divergence
+
+Assign each observed difference one classification:
+
+- **Required specialization:** the artifact or reference contract requires the
+  difference. No harmonization is recommended.
+- **Permitted variation:** implementations differ without meaningful
+  maintenance or correctness cost.
+- **Convention drift:** equivalent responsibilities use different names,
+  layouts, error behavior, metadata, validation, or tests without a reason.
+- **Architectural divergence:** responsibility is placed in the wrong layer or
+  bypasses a shared contract.
+- **Undetermined:** more evidence or a policy decision is needed.
+
+For required specialization, document the reason briefly so a future audit
+does not repeatedly flag it. For drift or architectural divergence, identify
+the preferred convention and why it is preferable.
+
+### 3. Build a duplication inventory
+
+Search for three kinds of repetition:
+
+1. **Exact duplication:** identical Go logic, prompt text, schema fragments, or
+   test support.
+2. **Structural duplication:** the same algorithm or lifecycle expressed with
+   renamed domain types.
+3. **Policy duplication:** the same invariant is independently encoded in
+   several production or test layers.
+
+Trace callers and consumers before recommending extraction. Record:
+
+- the repeated responsibility;
+- participating modules;
+- meaningful semantic differences;
+- change history or likely change cadence when discoverable;
+- defect risk if copies drift;
+- proposed owner and API shape; and
+- code or assets that would remain module-owned.
+
+Similar code is not sufficient evidence. Prefer a shared abstraction when at
+least one of the following is true:
+
+- three or more modules independently implement the same nontrivial invariant;
+- two modules share correctness-sensitive behavior that must evolve together;
+- existing duplicated code has already drifted or caused a defect; or
+- an existing shared contract is being reimplemented locally.
+
+Avoid extraction when the shared API would need artifact-specific callbacks,
+large configuration objects, type erasure, or branching on module identity.
+Those are signs that visual similarity is masking separate responsibilities.
+
+## Convention Review
+
+Evaluate the following conventions across all five modules.
+
+### Package organization
+
+- Comparable responsibilities use predictable filenames and package-local
+  ownership.
+- Optional specialized files, such as catalog or registry wiring, are present
+  only where the module has that responsibility.
+- Exported identifiers are limited to framework and registration contracts.
+- Test helpers remain local unless sharing them improves test clarity without
+  coupling independent suites.
+
+### Construction and registration
+
+- Required dependencies fail during construction.
+- Options are decoded and unknown options rejected consistently.
+- `ModuleSpec`, reference slots, artifact kind, execution class, and builder
+  behavior agree with runtime behavior.
+- Construction resolves stable reference-derived state where possible, while
+  operation requests own genuinely run- or chunk-specific inputs.
+- Returned specs, metadata, fingerprints, and byte slices have consistent
+  defensive-copy behavior.
+
+### Prompt and schema boundary
+
+- Shared prompt messages come from canonical shared assets; package assets
+  contain only lane-specific wording.
+- Shared content is exactly identical across manifests and appears in the
+  documented cache-friendly order.
+- Stable messages and cache boundaries precede the variable transcript.
+- Declared prompt inputs match reference slots and generated projections.
+- Private response schemas are strict structural envelopes and reject unknown
+  fields.
+- Semantic validation remains in deterministic code at the intended boundary.
+- Schema identity, version, digest, and diagnostics are exposed consistently
+  without leaking schema or prompt content.
+
+### Extraction and evidence
+
+- Request validation and contextual error wrapping follow one recognizable
+  pattern.
+- Reference material aids disambiguation but never becomes transcript evidence.
+- Source unit IDs are resolved against document order rather than numeric
+  assumptions.
+- Mapping performs only the deterministic transformations owned by extraction.
+- Canonical source references, stable ordering, and exact deduplication use
+  consistent policies where artifact semantics agree.
+- Whole-chunk evidence in scene descriptions is treated as an intentional
+  specialization rather than forced through citation-oriented helpers.
+- Results do not share mutable state with model responses, references, or
+  requests.
+
+### Provenance and diagnostics
+
+- Prompt and response-schema identities and hashes appear consistently in
+  manifest metadata.
+- Checkpoint fingerprints cover every stable semantic input that could change
+  accepted output, without including credentials, paths, timestamps, or source
+  content.
+- Registry and catalog projections use canonical bytes and retain bounded
+  provenance.
+- Errors and diagnostics are contextual, bounded, and free of raw reference or
+  secret content.
+
+### Tests
+
+- Each important contract has one clear test owner.
+- Equivalent risks receive comparable coverage without requiring identical test
+  file layouts.
+- Schema tests cover required structure and unknown-field rejection without
+  duplicating every semantic validator case.
+- Extractor tests cover request validation, response mapping, evidence,
+  ordering, references, and provider failures at the narrowest stable boundary.
+- Prompt tests prepare real embedded assets and protect input placement,
+  ordering, cache boundaries, and content-safety properties without freezing
+  exact prompt text.
+- Registration or production tests prove assembly once and do not repeat
+  package-local behavior unnecessarily.
+- Obsolete, redundant, representation-specific, or impossible fixtures are
+  identified for deletion or simplification.
+
+## Shared-Code Decision Framework
+
+Recommend the narrowest owner that matches the repeated responsibility:
+
+1. Keep artifact semantics in the module that owns the artifact.
+2. Use `internal/modules/dnd/shared` for D&D-wide mechanics with identical
+   semantics, such as prompt inputs or source-unit reference handling.
+3. Use a focused D&D subpackage for behavior shared by a subset of lanes, such
+   as NPC registry projection, when it has a coherent domain contract.
+4. Use a framework package only for transport-neutral or domain-neutral
+   behavior owned by the framework.
+
+Evaluate these candidate categories without assuming they should be extracted:
+
+- extractor request precondition validation;
+- private response-schema loading and metadata assembly;
+- prompt asset registration and hashing;
+- source-reference conversion, canonical ordering, and exact deduplication;
+- manifest metadata and checkpoint fingerprint assembly;
+- NPC registry resolution and names-only projection;
+- strict option decoding and module registration;
+- diagnostic redaction and bounded error context; and
+- repeated test fixtures or schema-validation utilities.
+
+For prompts and schemas:
+
+- Factor prompt wording into a shared asset only when every consumer needs the
+  exact same text and should receive future changes atomically.
+- Prefer an existing shared asset over a new near-duplicate.
+- Do not create shared prompt fragments solely because prose is similar.
+- Keep private response schemas package-owned unless a genuine shared wire
+  contract exists.
+- Do not introduce shared JSON Schema fragments or generation unless the audit
+  demonstrates a maintenance problem that outweighs tooling and indirection.
+
+## Additional Quality Review
+
+Beyond consistency and duplication, inspect:
+
+- functions with high cognitive complexity or responsibilities that can be
+  separated without obscuring the extraction flow;
+- repeated linear scans inside loops, avoidable serialization, unnecessary
+  allocations, or per-chunk reconstruction of stable state;
+- hidden mutation, aliasing, or inconsistent clone boundaries;
+- fingerprint omissions that could permit stale checkpoint reuse;
+- prompt inputs or references repeated unnecessarily across messages;
+- unreachable defensive checks or validation performed redundantly at several
+  layers;
+- errors that lose module, field, source, or stage context;
+- stale documentation, fixtures, names, compatibility aliases, or comments;
+- public or package abstractions that have only one artificial consumer; and
+- opportunities to delete code after a shared helper replaces it.
+
+Performance recommendations must identify a plausible workload and complexity
+impact. Do not recommend micro-optimization without evidence.
+
+## Evidence Collection
+
+Perform the audit in this order:
+
+1. Read the architecture, testing, documentation, module, LLM, pipeline, and
+   relevant integration contracts.
+2. Build the five-module comparison matrix from definitions and assets.
+3. Use graph similarity only to identify candidates; read the complete
+   functions and trace their callers before classifying them.
+4. Compare prompt manifests and shared assets byte-for-byte, then compare local
+   prompt semantics.
+5. Compare private schemas structurally and map each field to its DTO, mapper,
+   durable artifact, and validator owner.
+6. Trace reference and fingerprint data from construction through the LLM
+   request and checkpoint identity.
+7. Review focused tests alongside the behavior they own.
+8. Run the existing focused and repository-wide validation commands to
+   distinguish current failures from maintainability observations.
+
+Do not modify production code, tests, prompts, schemas, examples, or
+current-behavior documentation during the audit.
+
+## Finding Standard
+
+Every reported finding must contain:
+
+- severity: high, medium, or low;
+- category: convention drift, duplication, architecture, correctness,
+  performance, testing, or documentation;
+- affected modules and exact file or symbol references;
+- observed behavior and the convention or invariant it is compared against;
+- concrete maintenance, correctness, cost, or security impact;
+- recommended target state and ownership;
+- why the recommendation is preferable to leaving the code separate; and
+- validation or migration considerations.
+
+Order findings by severity and impact, not by module. Separate confirmed
+findings from optional improvements. State explicitly when no issue is found in
+a comparison area.
+
+For each apparent duplication, the audit must choose one outcome:
+
+- extract now;
+- harmonize without sharing;
+- retain intentionally separate; or
+- defer pending a named missing requirement.
+
+Do not report speculative abstractions as findings. Record them, if useful, as
+rejected or deferred candidates with the reason.
+
+## Audit Deliverable
+
+The completed audit should provide:
+
+1. an executive conclusion addressing convention consistency, shared-code
+   opportunities, and overall code quality;
+2. the completed five-module comparison matrix;
+3. prioritized findings with evidence and recommendations;
+4. intentional differences that should be preserved;
+5. rejected or deferred sharing candidates and rationale;
+6. a proposed refactoring sequence grouped into independently safe changes;
+   and
+7. validation commands and any residual risks.
+
+The audit should be actionable enough to support a later decision-complete
+implementation plan, but it must not implement or silently commit any
+recommendation.
+
+## Completion Criteria
+
+The strategy has been followed when:
+
+- all five primary packages and their prompts, schemas, registration,
+  provenance, reference handling, mapping, and tests have been compared;
+- every divergence has a classification;
+- every repeated candidate has an ownership and keep/share decision;
+- architecture and testing-policy constraints are applied explicitly;
+- findings cite exact evidence and explain impact;
+- intentional specialization is documented alongside drift;
+- repository validation results are recorded; and
+- no production changes were made as part of the audit.
+
+## Audit Results
+
+Status: Complete
+
+### Executive Conclusion
+
+The five D&D extraction modules follow a coherent overall convention: all are
+typed, production-registered extractors with strict private response schemas,
+package-owned artifact mapping, bounded provenance, shared canonical prompt
+assets, defensive result ownership, and deterministic downstream validation.
+Their differences in catalog/registry state, reference slots, local prompt
+assets, whole-chunk versus cited evidence, and scene prose cleanup are explained
+by artifact semantics and should remain.
+
+The consistency question is therefore **mostly yes, with two material
+exceptions**. Spell artifact ordering and all four citation extractors'
+reference ordering assume numeric unit IDs instead of source-document order,
+and spell/NPC checkpoint identities omit stable mapping policies. Both can
+change durable results or reuse results produced under different semantics.
+Smaller drift exists in common request preflight, scene semantic-validation
+ownership, prompt cache-boundary coverage, and package surface conventions.
+
+The shared-code question is **yes, but only for two narrow D&D-wide
+responsibilities**: common chunk-extraction preflight and document-aware
+source-reference ordering/canonicalization. They belong in
+`internal/modules/dnd/shared`, not in the framework. Private response DTOs,
+artifact mapping, structured LLM calls, prompt prose, response schemas,
+metadata assembly, error prefixes, and typed test fakes should remain
+package-owned; sharing them would require callbacks, type erasure, module
+branching, or broad configuration.
+
+Overall code quality is **good**. Construction, registration, cloning,
+diagnostics, schema loading, prompt composition, and validator composition are
+clear and consistently tested. The repository passes all current validation.
+The recommended work is targeted correction and consolidation, not a redesign
+of the extractor family.
+
+### Final Module Comparison Matrix
+
+| Module | Common contract | Required specialization | Prompt/schema boundary | Evidence and ordering | Provenance and tests |
+| --- | --- | --- | --- | --- | --- |
+| Spells | `dnd/spells` -> `dnd/spell-list`; typed LLM extractor; strict private DTO/schema; append merge and typed normalize/validate ([spec](../../internal/modules/dnd/extract/spells/extractor.go#L207)) | Prepared spell catalog and optional NPC registry projection; catalog prompt input ([constructor](../../internal/modules/dnd/extract/spells/extractor.go#L73)) | Shared evidence/input messages plus local task, instructions, catalog, and transcript; semantic catalog validation remains deterministic ([manifest](../../internal/modules/dnd/extract/spells/assets/prompts/dnd.spells.yaml#L23)) | Attaches current transcript ID and preserves invalid candidates, but sorts artifacts/references by numeric unit ID rather than document position ([canonicalization](../../internal/modules/dnd/extract/spells/canonicalize.go#L10)) | Prompt/schema/catalog/NPC projection fingerprints, but no mapping-policy fingerprint; broad extractor tests, incomplete full prompt cache-order assertion ([fingerprints](../../internal/modules/dnd/extract/spells/extractor.go#L144)) |
+| NPCs | `dnd/npcs` -> `dnd/npc-list`; same typed lifecycle ([spec](../../internal/modules/dnd/extract/npcs/extractor.go#L159)) | Deterministic NPC identity derivation; campaign references remain request context ([mapping](../../internal/modules/dnd/extract/npcs/canonicalize.go#L94)) | Shared evidence/input messages plus NPC-local task/instructions/transcript; private schema owns transport only ([manifest](../../internal/modules/dnd/extract/npcs/assets/prompts/dnd.npcs.yaml#L17)) | Artifact order resolves document positions; reference list still sorts numeric IDs before exact deduplication ([canonicalization](../../internal/modules/dnd/extract/npcs/canonicalize.go#L11)) | Prompt/schema/identity fingerprints, but no mapping-policy fingerprint; weakest focused preflight matrix and partial prompt ordering coverage ([fingerprints](../../internal/modules/dnd/extract/npcs/extractor.go#L91)) |
+| Combat turns | `dnd/combat-turns` -> `dnd/combat-turn-list`; same typed lifecycle ([spec](../../internal/modules/dnd/extract/combatturns/extractor.go#L201)) | Optional prepared NPC registry; all combat kinds remain raw validator candidates ([constructor](../../internal/modules/dnd/extract/combatturns/extractor.go#L68)) | Shared evidence/NPC/input messages plus combat-local task/instructions/transcript ([manifest](../../internal/modules/dnd/extract/combatturns/assets/prompts/dnd.combat_turns.yaml#L20)) | Artifact order uses valid source positions; extractor, normalizer, and invariant validator duplicate differing reference-order mechanics ([extractor](../../internal/modules/dnd/extract/combatturns/canonicalize.go#L10), [normalizer](../../internal/modules/dnd/normalize/combatturns/normalizer.go#L228)) | Mapping policy is fingerprinted; complete request preconditions, mapping, registry, and provider tests, but incomplete full prompt cache-order assertion ([fingerprints](../../internal/modules/dnd/extract/combatturns/extractor.go#L126)) |
+| NPC interactions | `dnd/npc-interactions` -> `dnd/npc-interaction-list`; same typed lifecycle ([spec](../../internal/modules/dnd/extract/npcinteractions/extractor.go#L203)) | Required NPC registry and names-only projection; exact identity remains in focused domain package ([constructor](../../internal/modules/dnd/extract/npcinteractions/extractor.go#L68)) | Shared evidence/NPC/input messages plus interaction-local task/instructions/transcript ([manifest](../../internal/modules/dnd/extract/npcinteractions/assets/prompts/dnd.npc_interactions.yaml#L20)) | Extractor references sort numeric IDs, while the interaction model already owns correct document-aware canonicalization at too-narrow a layer ([extractor](../../internal/modules/dnd/extract/npcinteractions/canonicalize.go#L10), [model helper](../../internal/modules/dnd/npcinteractions/canonical.go#L17)) | Mapping, identity, prompt, schema, and registry projection are fingerprinted; focused suite omits nil-receiver and full prompt cache-order cases ([fingerprints](../../internal/modules/dnd/extract/npcinteractions/extractor.go#L126)) |
+| Scene descriptions | `dnd/scene-descriptions` -> `dnd/scene-description-list`; same typed lifecycle ([spec](../../internal/modules/dnd/extract/scenedescriptions/extractor.go#L174)) | One summary per chunk, whole-chunk evidence, and fingerprinted prose trimming; no catalog/NPC projection ([mapping](../../internal/modules/dnd/extract/scenedescriptions/extractor.go#L150)) | Local task/instructions plus shared input/transcript; schema currently duplicates semantic enum/non-empty policy owned by shape validation ([schema](../../internal/modules/dnd/extract/scenedescriptions/assets/schemas/dnd_scene_descriptions_llm.v1.json#L8), [validator](../../internal/modules/dnd/validate/scenedescriptions/shape/validator.go#L49)) | Whole materialized chunk becomes one source range; citation ordering is not applicable | Mapping policy is fingerprinted; strongest full prompt order/cache test, but preflight omits nil receiver/context cases ([fingerprints](../../internal/modules/dnd/extract/scenedescriptions/extractor.go#L98), [prompt test](../../internal/modules/dnd/extract/scenedescriptions/scriptorium_assets_test.go#L15)) |
+
+All five reject unknown options, register through the production D&D registrar,
+return independently owned results, and use package-local structured response
+types. No inconsistent secret handling, raw prompt/schema diagnostic exposure,
+hidden result aliasing, or unregistered audited extractor was found.
+
+### Prioritized Findings
+
+#### High
+
+1. **Source references and spell artifacts can be durably ordered contrary to
+   transcript order.**
+   - **Category:** correctness and architecture.
+   - **Affected modules:** spell, NPC, combat-turn, and NPC-interaction
+     extractors; related spell/combat/NPC normalizers and combat invariants.
+   - **Evidence:** spell selects the smallest positive numeric unit ID
+     ([`earliestSourceUnit`](../../internal/modules/dnd/extract/spells/canonicalize.go#L62));
+     all four citation extractors sort reference endpoints numerically
+     ([spell](../../internal/modules/dnd/extract/spells/canonicalize.go#L30),
+     [NPC](../../internal/modules/dnd/extract/npcs/canonicalize.go#L31),
+     [combat](../../internal/modules/dnd/extract/combatturns/canonicalize.go#L30),
+     [interaction](../../internal/modules/dnd/extract/npcinteractions/canonicalize.go#L30)).
+     Valid source documents require unique positive IDs, not monotonically
+     increasing IDs
+     ([`ValidateDocument`](../../internal/core/source/validation.go#L8)).
+     The interaction model demonstrates the correct document-aware comparison
+     ([`SourceRefLess`](../../internal/modules/dnd/npcinteractions/canonical.go#L51)).
+   - **Impact:** valid evidence can be reordered away from transcript order;
+     spell casts with invalid or later evidence can precede earlier valid
+     casts. This changes durable list order, evidence presentation, merge input,
+     and checkpointed results.
+   - **Target state and owner:** move a document-backed `SourceRefOrder` to
+     `internal/modules/dnd/shared`, with `EarliestValid` and `Canonicalize`
+     operations. Preserve invalid candidates, exact deduplication, stable ties,
+     nil/empty distinction, cloning, and deterministic invalid fallback.
+     Artifact comparison, DTO conversion, source-ID attachment, and repair
+     accounting remain local.
+   - **Why shared:** the rule is a D&D-wide evidence invariant already
+     implemented by extract, normalize, and validate consumers; continued
+     copies have already diverged. A shared position index also reduces repeated
+     `UnitIndex` scans from approximately `O(A log A * R * U)` ordering work to
+     `O(U + A*R + A log A)` for `A` artifacts, `R` references, and `U` units.
+   - **Migration/validation:** add non-monotonic, invalid, duplicate, nil/empty,
+     aliasing, stable-tie, and repair-count tests; migrate the interaction model
+     and combat normalize/invariant pair before extractors; add/bump mapping
+     policy fingerprints so old checkpoints miss intentionally.
+
+2. **Spell and NPC checkpoint identities omit stable mapping policies.**
+   - **Category:** correctness and data integrity.
+   - **Affected modules:** spell and NPC extractors.
+   - **Evidence:** both extractors perform deterministic ordering,
+     canonicalization, source attachment, and mapping
+     ([spell mapping](../../internal/modules/dnd/extract/spells/canonicalize.go#L10),
+     [NPC mapping](../../internal/modules/dnd/extract/npcs/canonicalize.go#L11)),
+     but their fingerprint providers name catalog/projection or identity
+     policies without a mapping policy
+     ([spell fingerprints](../../internal/modules/dnd/extract/spells/extractor.go#L144),
+     [NPC fingerprints](../../internal/modules/dnd/extract/npcs/extractor.go#L91)).
+     Prepared fingerprints are lane-scoped
+     ([collector](../../internal/framework/pipeline/prepared_fingerprints.go#L23))
+     and restore requires exact normalized equality
+     ([comparison](../../internal/framework/checkpoint/loader.go#L301)).
+   - **Impact:** a mapping-policy code change can reuse a checkpoint produced
+     under older artifact/evidence semantics when prompt and schema bytes are
+     unchanged.
+   - **Target state and owner:** each extractor owns an explicit stable
+     mapping/canonicalization policy fingerprint. Metadata assembly stays
+     package-local.
+   - **Why preferable:** a local named semantic fingerprint directly closes
+     the reuse gap; a generic metadata builder would only hide module-specific
+     omissions behind configuration.
+   - **Migration/validation:** add provider and prepared-checkpoint restore
+     tests. Adding a fingerprint safely invalidates existing identities by full
+     list mismatch; no artifact payload migration is required.
+
+#### Medium
+
+3. **Common extraction preflight is copied across five production callers and
+   has already drifted in test protection.**
+   - **Category:** duplication and testing.
+   - **Affected modules:** all five extractors.
+   - **Evidence:** every `Extract` validates the same context/source/chunk/unit
+     prerequisites before calling
+     [`ChunkPromptMaterial`](../../internal/modules/dnd/shared/extraction_inputs.go#L12),
+     but focused coverage ranges from a complete spell/combat matrix to only
+     cancellation and source mismatch for NPCs
+     ([spell tests](../../internal/modules/dnd/extract/spells/extractor_test.go#L235),
+     [NPC tests](../../internal/modules/dnd/extract/npcs/extractor_test.go#L139)).
+   - **Impact:** validation order, error behavior, or a newly required common
+     precondition can diverge silently among modules.
+   - **Target state and owner:** add
+     `shared.PrepareChunkExtraction(ctx, req) (contracts.LLMInputMaterial,
+     error)` for cancellation, non-nil source/chunk, non-empty units, and
+     matching cloned material. Receiver/client checks, specialized references,
+     provider calls, typed results, and contextual wrapping remain local.
+   - **Why shared:** this extends the existing common material boundary with
+     one coherent invariant and serves five real callers without callbacks or
+     module configuration.
+   - **Migration/validation:** shared table tests own common inputs; package
+     tests retain nil receiver/dependency, specialized reference, wrapped
+     error, and provider cases. Preserve current validation order and useful
+     package context.
+
+4. **Scene semantic validity has two production owners.**
+   - **Category:** architecture.
+   - **Affected module:** scene descriptions.
+   - **Evidence:** the private response schema enforces the scene-kind enum and
+     non-empty title/summary
+     ([schema](../../internal/modules/dnd/extract/scenedescriptions/assets/schemas/dnd_scene_descriptions_llm.v1.json#L8));
+     deterministic shape validation independently enforces the same policy
+     ([validator](../../internal/modules/dnd/validate/scenedescriptions/shape/validator.go#L49)).
+     The documented schema boundary assigns semantic enum/non-empty rules to
+     deterministic validators
+     ([LLM internals](../internal/llm.md#L169)).
+   - **Impact:** the two policies can drift and produce provider-dependent
+     rejection before typed validation, while non-LLM artifacts see only the
+     validator.
+   - **Target state and owner:** retain JSON type, required/nullability, and
+     unknown-field constraints in the package-private schema; make the scene
+     shape validator the sole semantic owner.
+   - **Why harmonize without sharing:** only scene has this duplicated policy;
+     moving it to the existing validator removes an owner without inventing an
+     abstraction.
+   - **Migration/validation:** update schema structural tests and validator
+     semantic tests, then expect the schema digest/checkpoint identity to
+     change. Verify provider-decoded invalid candidates reach deterministic
+     validation.
+
+#### Low
+
+5. **Four prompt suites do not fully protect the documented message-order and
+   cache-boundary contract.**
+   - **Category:** testing.
+   - **Affected modules:** spells, NPCs, combat turns, and NPC interactions.
+   - **Evidence:** scene descriptions asserts the complete prepared role order
+     and ephemeral/no-cache placement
+     ([test](../../internal/modules/dnd/extract/scenedescriptions/scriptorium_assets_test.go#L15));
+     the four citation suites cover registration and selected inputs but not
+     the full documented sequence
+     ([spell tests](../../internal/modules/dnd/extract/spells/scriptorium_assets_test.go),
+     [NPC tests](../../internal/modules/dnd/extract/npcs/scriptorium_assets_test.go),
+     [combat tests](../../internal/modules/dnd/extract/combatturns/scriptorium_assets_test.go),
+     [interaction tests](../../internal/modules/dnd/extract/npcinteractions/scriptorium_assets_test.go)).
+   - **Impact:** manifest edits can move variable content into a cacheable
+     prefix or reorder stable grounding without a focused failure, increasing
+     request cost or reducing prompt quality.
+   - **Target state and owner:** each package test should assert its complete
+     documented prepared sequence and cache flags using the real registry.
+   - **Why local:** lane inputs differ and the behavior belongs to each prompt
+     manifest; shared test setup would obscure the boundary.
+   - **Migration/validation:** assert roles/input identities/cache flags, not
+     exact prompt text, byte counts, or hashes.
+
+6. **Small exported-surface and defensive-behavior drift remains.**
+   - **Category:** convention drift.
+   - **Affected modules:** spells, combat turns, and D&D shared.
+   - **Evidence:** spell and combat export unused singular `ArtifactType`
+     constants while production uses typed `ArtifactKind`
+     ([spell](../../internal/modules/dnd/extract/spells/extractor.go#L17),
+     [combat](../../internal/modules/dnd/extract/combatturns/extractor.go#L17));
+     spell `ManifestMetadata` lacks the nil guard used by the other four
+     ([spell](../../internal/modules/dnd/extract/spells/extractor.go#L122),
+     [NPC example](../../internal/modules/dnd/extract/npcs/extractor.go#L74));
+     [`SourceRefCandidate`](../../internal/modules/dnd/shared/unit_refs.go#L90)
+     has only a test caller, ignores its document parameter, and is unsafe for
+     extractor provenance because it trusts model-supplied source identity.
+   - **Impact:** the package surface presents competing artifact vocabulary,
+     zero-value behavior is inconsistent, and an artificial shared API invites
+     incorrect reuse. Immediate runtime impact is limited.
+   - **Target state and owner:** delete the unused constants; align spell nil
+     metadata behavior locally; delete `SourceRefCandidate` when the correct
+     shared reference API lands.
+   - **Why preferable:** deletion and local harmonization clarify existing
+     contracts without adding a helper.
+   - **Migration/validation:** graph search found no production consumers;
+     compile all internal packages and add one spell zero-value metadata test.
+
+#### Optional improvement
+
+The current module documentation lists only four production consumers of
+`ChunkPromptMaterial`, omitting scene descriptions
+([documentation](../internal/modules.md#L67)). Correct that list when the shared
+preflight boundary is documented. This is localized documentation maintenance,
+not a separate production design finding.
+
+No actionable issue was found in option strictness, production registration,
+typed codec selection, result cloning, prompt/schema content redaction, local
+prompt semantic ownership, catalog/NPC source-evidence separation, provider
+error wrapping, or secret handling.
+
+### Intentional Differences To Preserve
+
+- **Prepared specialized state:** spells retains a catalog and optional NPC
+  resolver; combat retains an optional NPC resolver; interactions requires one;
+  NPC and scene extraction need only request material. Construction-time
+  identity belongs only where the artifact uses it.
+- **Reference slots:** catalog and NPC registry slots follow grounding needs.
+  The interaction registry is required; combat/spell registry use is optional.
+  Campaign references remain prompt context and never become evidence.
+- **Evidence model:** scene descriptions cites the whole materialized chunk;
+  forcing it through citation DTO/canonicalization machinery would weaken its
+  one-summary-per-chunk contract.
+- **Prompt assets:** only byte-identical canonical messages are shared. Local
+  task, instruction, catalog, identity, and evidence wording changes with the
+  artifact and must not be coupled.
+- **Private schemas and DTOs:** similar `source_refs` fragments do not form a
+  separately versioned wire contract. Package ownership keeps response changes
+  aligned with mapping and diagnostics.
+- **Prose handling:** scene title/summary trimming is a deliberate fingerprinted
+  mapping policy; citation-lane names and enum candidates remain raw for their
+  normalizers and validators.
+- **Test organization:** file layout, typed fakes, schema compilation helpers,
+  and real-registry prompt setup remain package-local. Equivalent risks need
+  comparable coverage, not identical fixtures or filenames.
+
+### Shared-Code Decisions
+
+| Decision | Owner and scope | Why this is the narrow correct boundary |
+| --- | --- | --- |
+| Extract common preflight | `internal/modules/dnd/shared`; validate common typed extraction request state and return cloned matching `LLMInputMaterial` | Five current callers repeat one prerequisite to the existing `ChunkPromptMaterial` boundary. Receiver/dependency/specialized checks remain readable and local. |
+| Extract source-reference order | `internal/modules/dnd/shared`; document-backed `SourceRefOrder.EarliestValid` and `.Canonicalize` over durable `[]source.SourceRef` | Multiple extract, normalize, model, and validate consumers must evolve together; the rule is D&D-wide but has no demonstrated non-D&D/framework consumer. |
+| Harmonize mapping fingerprints | Spell and NPC extractor packages | The missing values are module-semantic; local named fingerprints are safer than a configurable metadata builder. |
+| Harmonize scene validation ownership | Private scene schema plus `validate/scenedescriptions/shape` | Remove semantic keywords from the transport schema and keep the already registered durable validator as sole owner. No sharing is needed. |
+| Harmonize prompt contract tests | Four citation extractor test suites | Each manifest owns its role/input/cache sequence; consistent assertions should stay beside distinct assets. |
+| Harmonize preflight coverage | All five package suites after shared preflight | Shared tests own common branches; local tests retain package-visible context and specializations. |
+| Harmonize package hygiene | Spell/combat extractors and D&D shared | Delete unused exports/artificial API and align spell metadata nil behavior without introducing a new abstraction. |
+
+### Rejected Or Deferred Candidates
+
+| Candidate | Decision and rationale |
+| --- | --- |
+| Deprecated `roster` slot asymmetry | **Deferred pending a product compatibility decision.** Configuration documents `roster` as a deprecated `party` alias ([config](../config.md#L526)), while scene removes it. Choose a removal release or uniform alias lifetime before changing slots and migration guidance. |
+| Spell response-schema key/ID naming | **Deferred pending a provenance migration decision.** The spell names differ from the `_llm`/`.llm` convention, but schema identity is persisted metadata. Renaming needs an alias or an explicit manifest/checkpoint compatibility break. |
+| Prompt registration/hash wrapper | **Rejected.** `shared.PromptAssetManifest` already owns coherent composition/hashing. Hiding filesystems, registries, `sync.Once`, and diagnostic nouns would require broad configuration. |
+| Manifest/fingerprint builder | **Rejected.** Stable inputs are module-specific; a generic builder would accept the same keys/policies as arguments and could conceal omissions such as the current spell/NPC gap. |
+| Generic artifact mapper | **Rejected.** Private DTO fields, identity, enums, and artifact types require callbacks, type erasure, or module branching. Only reference ordering is genuinely common. |
+| Generic structured-LLM call wrapper | **Rejected.** Generic output and request configuration would hide an already clear framework client boundary without removing semantic work. |
+| Shared error-prefix helper | **Rejected.** It would add a module-name parameter to replace three transparent lines and weaken local diagnostic ownership. |
+| Shared response schemas or prompt prose | **Retain intentionally separate.** Similar structure/text is not an atomic shared contract; private schema and lane prompt changes should not propagate together. |
+| Shared schema-test or provider-fake utilities | **Retain intentionally separate.** Typed DTOs and package assets are the behavior under test; central fixtures would couple suites and obscure failures. |
+| Framework-level preflight/reference API | **Rejected.** No non-D&D consumer or framework-owned invariant was found. The D&D shared layer preserves dependency direction and domain ownership. |
+
+### Recommended Refactoring Sequence
+
+1. **Land checkpoint identity protection independently.** Add spell and NPC
+   mapping-policy fingerprints with provider and checkpoint-restore tests.
+   This is a small package-local change and protects all later semantic
+   migrations from stale reuse.
+2. **Consolidate source-reference mechanics.** Add and exhaustively test the
+   D&D shared document-position index/API. Migrate the existing interaction
+   model helper first, then combat normalize plus invariants, then NPC/spell
+   normalizers, and finally the four citation extractors. Keep package mapping
+   and repair accounting local; remove numeric comparators and the unused
+   `SourceRefCandidate` only after all consumers move.
+3. **Consolidate common extraction preflight.** Extend the existing shared
+   chunk-material boundary, migrate one extractor to establish error/validation
+   compatibility, then migrate the remaining four. Move common branch tables
+   to shared tests and retain package-context smoke tests.
+4. **Resolve scene validation ownership independently.** Adjust schema tests,
+   remove semantic enum/non-empty constraints from the private schema, retain
+   validator cases, and verify the expected schema fingerprint/checkpoint miss.
+5. **Close prompt contract coverage.** Add full role/input/cache-boundary
+   assertions to each citation prompt suite without shared fixtures or exact
+   text/hash assertions.
+6. **Apply low-risk package/documentation cleanup.** Remove unused
+   `ArtifactType` constants, add the spell metadata nil guard/test, and update
+   module documentation for the shared helper consumers and boundaries.
+
+Each scope can be reviewed and reverted independently. The sequence is not a
+decision-complete implementation plan; implementation should still pin exact
+fingerprint values, exported names, error compatibility, and per-package test
+cases.
+
+### Validation And Residual Risks
+
+Final validation on 2026-07-24:
+
+```text
+go test -count=1 ./...   PASS
+go vet ./...             PASS
+go build ./cmd/notarius  PASS
+gofmt -l .               PASS (no files listed)
+git diff --check         PASS
+git diff --no-index --check /dev/null docs/roadmap/audit.md
+                         PASS (no whitespace errors; exit 1 denotes differences)
+```
+
+The audit also checked that cited relative paths exist in the current working
+tree. The audit changed no production code, tests, prompts, schemas, examples,
+or current-behavior documentation.
+
+Residual risks and limits:
+
+- This was a static and deterministic-test audit; it did not evaluate live
+  model extraction quality, prompt effectiveness, provider cache-hit rates, or
+  token cost. Prompt changes still need representative human/model evaluation.
+- Existing tests use mostly monotonic unit IDs, so the ordering defect is not a
+  current failing test. The proposed non-monotonic fixtures are required before
+  changing behavior.
+- The exact durable order expected for invalid references must remain
+  deterministic and diagnostics-friendly during API design; invalid references
+  must not be discarded merely because default validators usually reject them
+  later.
+- Mapping and schema fingerprint additions intentionally invalidate prior
+  checkpoint identities. Operators should be told to expect recomputation; no
+  serialized artifact migration is otherwise indicated.
+- The `roster` alias lifetime and spell schema-identity migration remain human
+  product/compatibility decisions.
+- Passing repository checks establishes current deterministic correctness, not
+  absence of model-quality regressions or correctness under unrepresented
+  source-document shapes.

docs/roadmap/dnd-npc-interactions.md

@@ -0,0 +1,200 @@
+# D&D NPC Interactions
+
+Status: Implemented
+
+The current durable and configuration contract is documented in the
+[D&D NPC interaction artifact](../integrations/dnd-npc-interaction-artifacts.md).
+
+## Purpose
+
+The normalized NPC registry intentionally answers only who was identified in a
+session. It does not answer whether an NPC was merely mentioned, participated
+in dialogue, or fought alongside or against the party. Add a separate,
+ordered NPC-interaction artifact for that occurrence-level information rather
+than expanding the identity registry.
+
+This feature preserves the minimal-extractor policy: the model identifies one
+bounded kind of interaction and its supporting transcript evidence. It does
+not summarize the interaction, infer relationships, or maintain NPC state.
+
+## Desired End State
+
+A D&D pipeline can run NPC extraction first and supply its accepted normalized
+registry to a later `dnd/npc-interactions` lane. The later lane emits an ordered
+list of evidenced interaction occurrences involving registry NPCs.
+
+The production identities should be:
+
+- extractor and normalizer key: `dnd/npc-interactions`;
+- artifact kind: `dnd/npc-interaction-list`;
+- durable schema ID: `notarius.dnd.npc_interactions`;
+- durable schema name: `notarius_dnd_npc_interactions_v1`;
+- durable schema version: `v1`; and
+- media type: `application/json`.
+
+The lane must use the existing D&D module organization, shared prompt assets,
+typed artifact pipeline, codec boundary, registration pattern, and default
+validator composition.
+
+## Artifact Contract
+
+The durable payload is an object containing an `interactions` array. The array
+may be empty. Each interaction contains exactly:
+
+- `name`: the canonical NPC name from the supplied registry;
+- `kind`: one value from the bounded interaction vocabulary; and
+- `source_refs`: one or more current-transcript ranges supporting both the NPC
+  identity and the classified interaction.
+
+Every object rejects unknown fields. The model-facing response should contain
+only the corresponding name, kind, and source-unit range candidates. Notarius
+attaches the current source identity deterministically; the model must not
+reproduce it.
+
+Do not add a separate interaction ID in the durable contract. Stable ordering,
+the canonical NPC name, the bounded kind, and exact evidence ranges are enough
+to identify and audit an occurrence for the present use cases. Revisit durable
+cross-artifact identity only with a concrete consumer requirement.
+
+## Interaction Vocabulary
+
+Use this closed vocabulary:
+
+| Kind | Meaning |
+| --- | --- |
+| `mentioned` | The NPC is referred to, but is not established as present or communicating in the evidenced passage. |
+| `noncombat_presence` | The NPC is present and relevant to the passage but does not meaningfully participate in dialogue or combat. |
+| `dialogue` | The NPC speaks, responds, or is directly engaged in a meaningful non-combat exchange. |
+| `combat_ally` | The NPC actively participates in combat on the party's side. |
+| `combat_opponent` | The NPC actively participates in combat against the party. |
+| `other` | The transcript clearly establishes a direct NPC occurrence that fits none of the preceding kinds. |
+
+`other` is a residual category for positively evidenced activity, not an
+escape hatch for uncertain classification. Omit a candidate when the
+transcript does not support one category.
+
+When activities overlap within one occurrence, apply this precedence:
+
+1. active combat participation outranks dialogue, presence, and mention;
+2. dialogue outranks non-combat presence and mention;
+3. non-combat presence outranks mention; and
+4. `other` applies only when none of the defined categories describes the
+   evidenced activity.
+
+Combat alignment is not resolved by precedence. An NPC cannot be both a combat
+ally and combat opponent in one occurrence; split the record when its alignment
+meaningfully changes.
+
+## Occurrence Boundaries And Ordering
+
+An occurrence represents one NPC, one interaction kind, and one locally
+coherent transcript passage. Combine repeated evidence only while it supports
+the same uninterrupted activity. Create separate occurrences when:
+
+- the interaction kind changes;
+- combat alignment changes;
+- a scene or meaningful absence separates repeated activity; or
+- the NPC is first mentioned and later becomes present.
+
+Extraction remains chunk-scoped, so an occurrence must not span accepted chunk
+boundaries. Merge and normalization must not semantically combine occurrences
+from different chunks.
+
+Order the durable list by the earliest valid source position. Use canonical
+name, interaction kind, and canonical source ranges as deterministic
+tie-breakers. Remove only exact duplicate records; do not use model judgment to
+collapse nearby occurrences.
+
+## NPC Grounding And Evidence
+
+The extractor declares the existing `npcs` reference slot and requires a bound,
+accepted NPC registry. The normal same-run configuration binds the normalized
+output of an earlier NPC lane through an ordered generated reference. Existing
+framework support for a compatible external registry may remain available, but
+must not weaken the artifact contract.
+
+Present the registry's names-only projection to the model. An emitted name must
+match one canonical registry name exactly after the registry's established
+lookup rules are applied, and the durable artifact retains that canonical
+display name. Do not copy NPC IDs or registry source references into an
+interaction.
+
+The registry establishes available identity, not occurrence. Every interaction
+must cite current transcript ranges that independently support both the NPC and
+the selected kind. Campaign references and generated artifacts may disambiguate
+a name, but they never become interaction evidence.
+
+If no registry NPC has an evidenced interaction, the correct artifact is an
+empty list. A missing, rejected, or incompatible required registry handoff is a
+pipeline dependency failure rather than a request to extract ungrounded names.
+
+## Prompt And Model Boundary
+
+Follow the established D&D prompt ordering and cache-boundary policy. Stable
+shared instructions, lane instructions, campaign references, and the NPC
+names-only projection precede the variable transcript. Factor wording shared
+with the spell and combat lanes into the existing shared asset pattern rather
+than creating nearly identical package-local messages.
+
+The prompt must:
+
+- define the closed vocabulary and precedence rules;
+- ask for occurrences involving only supplied registry NPCs;
+- distinguish mention from presence and dialogue;
+- require current-transcript evidence for identity and kind;
+- forbid summaries, relationship inference, sentiment, aliases, and invented
+  names; and
+- permit an empty result.
+
+Keep the private JSON Schema structural. Deterministic code owns canonical-name
+resolution, enum enforcement at the durable boundary, source-range validation,
+ordering, exact deduplication, and evidence invariants.
+
+## Validation And Quality
+
+Provide production validators and default chains at extraction and
+normalization boundaries consistent with the existing D&D artifacts. The
+append-only merge does not require a separate default validator chain.
+Deterministic validation must reject:
+
+- missing or extra fields;
+- empty or unrecognized NPC names;
+- names absent from the supplied registry;
+- unknown interaction kinds;
+- empty, malformed, out-of-source, or reversed evidence ranges; and
+- records whose evidence comes from a reference rather than the current
+  transcript.
+
+Any relatedness validator should remain warning-only unless evaluation
+demonstrates a reliable deterministic rejection rule. Diagnostics must be
+bounded and must not leak reference contents.
+
+Evaluate the lane on a small human-reviewed transcript set that includes every
+category, transitions between categories, multiple occurrences for one NPC,
+mentions followed by appearances, alignment changes, repeated evidence across
+chunks, and empty output. Review category agreement, evidence sufficiency,
+ordering, duplicate behavior, and reliability on the smaller models the
+application is intended to support. Treat model-output evaluation as a human
+development aid, not a brittle deterministic test oracle.
+
+## Documentation Outcomes
+
+When implemented, document the durable artifact in `docs/integrations/`, add
+the selectable module and validator contracts to configuration documentation,
+update the current module and pipeline internals, and provide a maintained
+ordered-pipeline example showing NPC extraction followed by interaction
+extraction. Future behavior must remain in this roadmap until it exists.
+
+## Non-Goals
+
+This scope does not:
+
+- add occurrence fields to the normalized NPC registry;
+- summarize dialogue, combat, or NPC behavior;
+- infer disposition, relationships, factions, motives, or persistent state;
+- identify player characters or anonymous groups as NPCs;
+- add scene participants or duplicate scene-description responsibilities;
+- reconcile NPC aliases or perform LLM-assisted semantic deduplication;
+- derive interaction records from registry evidence; or
+- introduce a DAG, concurrent cross-lane reconciliation, or a new reference
+  mechanism.

docs/roadmap/future.md

@@ -16,28 +16,18 @@ not as committed release dates.
   validator, and normalizer development. Treat model-quality review as an
   iterative human evaluation aid, not a deterministic correctness gate.
 
-### Expand Sequential D&D Artifacts
+### Use D&D Scene Chunking
 
-- Add narrative extraction for scene summaries, party actions, and NPCs
-  encountered when that output proves useful beyond the dedicated NPC artifact.
-- Use ordered pipeline steps when a later artifact needs an accepted earlier
-  artifact as context. Keep independent lanes in the same step and do not
-  introduce a general DAG or concurrent cross-lane reconciliation model.
-
-### Improve D&D Scene Classification
-
-- Extend scene annotations with classifications that downstream extractors can
-  use, including reliable combat and narrative indicators.
-- Strengthen the scene prompt so every scene containing combat turns is marked
-  as combat, and add validation capable of detecting missing or inconsistent
-  combat classifications.
-- Allow the combat extractor to no-op for chunks that are not classified as
-  combat, avoiding unnecessary model calls where practical.
-- Allow a narrative extractor to select the corresponding scene classification
-  rather than processing every chunk indiscriminately.
+- Allow the combat extractor to no-op for chunks classified as non-combat only
+  after the scene-description artifact can be supplied through an explicit
+  ordered dependency. Do not make generic chunk materialization depend on a D&D
+  classification.
+- Use ordered pipeline steps whenever a later artifact needs an accepted
+  earlier artifact as context. Keep independent lanes in the same step and do
+  not introduce a general DAG or concurrent cross-lane reconciliation model.
 - Reassess whether one shared scene plan provides enough context for NPC,
-  spell, combat, and narrative pipelines after these extractors have real-world
-  usage. Add more complex chunking only in response to demonstrated failures.
+  spell, combat, interaction, and scene-description lanes after real-world use.
+  Add more complex chunking only in response to demonstrated failures.
 
 ## Shared Normalization And Quality Work
 

docs/roadmap/implementation.md

@@ -1,373 +1,425 @@
-# Minimal D&D Extraction Contracts Implementation Plan
+# D&D Extraction Module Audit Execution Plan
 
-**Status:** Implemented
+Status: Ready for audit execution
 
-## Objective
+This document divides the audit defined in
+[D&D Extraction Module Refactoring Audit Strategy](audit.md) into five prompts.
+Execute the stages in order. Each stage is a read-only code audit: it may update
+`docs/roadmap/audit.md`, but it must not change production code, tests, prompts,
+schemas, examples, current-behavior documentation, or configuration.
 
-Implement the durable contract redesign defined by
-[Minimal D&D Extraction Contracts](minimal-dnd-extraction-contracts.md) and
-[ADR-0009](../adr/0009-minimal-evidence-grounded-extraction-artifacts.md).
-The result is a coordinated in-place redesign of NPC, spell-cast, and combat-turn
-extraction. Each lane must emit only its narrow, evidence-grounded facts; no
-removed rich-schema field may survive as an optional field, placeholder, compatibility
-shim, or deterministic synthesis.
+The word "implementation" in this filename refers to implementing the audit
+strategy, not implementing refactors.
 
-This plan deliberately starts with NPCs because the NPC registry is a generated
-reference consumed by spell and combat extraction. Spell and combat then change
-independently, followed by one repository-wide contract and documentation
-pass.
+## Global Instructions
 
-## Fixed Decisions And Guardrails
+Every stage must:
 
-The implementing agent must treat these as decisions, not open design choices:
+1. Read `docs/roadmap/audit.md` in full before beginning.
+2. Follow `AGENTS.md`, `docs/development.md`, and all policies under
+   `docs/policy/`.
+3. Prefer the codebase knowledge graph for code discovery and call tracing.
+   Use direct file and text searches for prompt assets, JSON Schemas,
+   configuration, documentation, and exact string comparison.
+4. Inspect all five extractors where the stage calls for comparison:
+   `spells`, `npcs`, `combatturns`, `npcinteractions`, and
+   `scenedescriptions`.
+5. Cite exact files and symbols for every observation written to `audit.md`.
+6. Distinguish confirmed facts, provisional interpretations, intentional
+   specialization, and actionable findings.
+7. Avoid treating visual similarity, file count, or line count as sufficient
+   evidence for sharing.
+8. Preserve findings from earlier stages unless new evidence disproves them.
+   When revising an earlier conclusion, edit it in place and record the reason;
+   do not append contradictory conclusions.
+9. Keep implementation out of scope. Do not modify code to test whether a
+   proposed refactor is convenient.
+10. Leave the repository otherwise unchanged and report any pre-existing dirty
+    worktree state before proceeding.
 
-- Keep the artifact kinds `dnd/npc-list`, `dnd/spell-list`, and
-  `dnd/combat-turn-list`, their module keys and capabilities, their media types,
-  and their prompt IDs stable.
-- Retain exactly the existing v1 schema keys, IDs, names, versions, filenames,
-  prompt IDs, and prompt versions listed in the feature roadmap. Change their
-  unpublished shapes and content in place; do not harmonize the private spell
-  identity as part of this work.
-- Do not add v2 assets, runtime version negotiation, migration code, dual-write
-  behavior, a second module registration, or compatibility fixtures for the
-  superseded pre-release shapes. Existing fixtures should be rewritten or
-  deleted according to whether they still protect current behavior.
-- Require the top-level list field and every record field. A list may be empty,
-  but every returned record must have at least one source reference. Use strict
-  JSON objects with unknown fields rejected and no nullable or optional legacy
-  fields.
-- Keep private source references limited to `start_unit_id` and `end_unit_id`.
-  Mapping assigns the current source document ID; campaign references and an
-  NPC registry never become event evidence.
-- Keep the shared prompt ordering policy: stable shared instructions first,
-  stable campaign and generated references next, module task material after
-  those references, and the chunk-variable transcript last.
-- Preserve the existing production validator-chain order and the rule that a
-  configured chain is authoritative. Remove or simplify validators; do not
-  silently reorder chains.
-- Preserve the existing NPC ID derivation algorithm and comparison/display
-  normalization for canonical names. Retain existing semantic-policy
-  identifiers as well as schema and prompt versions; pre-redesign development
-  state is disposable. Unchanged canonical names must not receive new IDs.
-- Use deterministic code only for display normalization, known catalog or NPC
-  canonicalization, application-owned IDs, evidence canonicalization, ordering,
-  and exact duplicate collapse. Do not introduce fuzzy matching or inferred
-  enrichment.
-- Follow the testing policy: protect schemas and behavior at their owning
-  boundaries, delete obsolete tests, and avoid tests that snapshot prompt prose
-  or detect exact shared-prefix length. No live or paid model call is a stage
-  completion requirement.
-- Each stage must leave the repository compiling and `go test ./...` passing.
-  Update all affected fixtures and callers within the stage that changes a
-  public Go type; do not leave an intentionally broken intermediate commit.
+## Working Results Structure
 
-## Stage 1: Cut Over The NPC Contract And Name Projection
+Stages 1 through 4 maintain one working area at the end of `audit.md`:
 
-This stage establishes the redesigned generated reference on which later stages
-depend.
+```markdown
+## Audit Results
 
-### 1.1 Replace the durable and private data shapes
+Status: In progress
 
-- Reduce `dnd.NPC` in `internal/modules/dnd/types.go` to exactly `ID`, `Name`,
-  and `SourceRefs`. Delete `NPCRelationship` and every alias, description, and
-  relationship field or helper that becomes unused.
-- Rewrite the existing strict durable schema `dnd_npcs.v1.json` with required
-  top-level `npcs`, and records containing only required `id`, `name`, and
-  `source_refs`. Keep the NPC codec at schema version `v1`, name
-  `notarius_dnd_npcs_v1`, and its existing schema ID and media type.
-- Rewrite the strict private schema `dnd_npcs_llm.v1.json`. Its records contain
-  only required `name` and model-facing source ranges. Keep the extractor schema
-  identity and prompt ID/version unchanged.
-- Reduce the private response DTO and mapping accordingly. Preserve candidate
-  data at the mapping boundary so deterministic validators, rather than mapping
-  defaults, reject blank names or invalid ranges. Assign the current source ID
-  and derive the NPC ID in application code.
-- Rewrite the NPC task and instruction assets to ask only for individually
-  identifiable NPC names and supporting transcript ranges. Explicitly exclude
-  anonymous groups, generic roles, invented labels, descriptions, aliases, and
-  relationships. Keep transcript material last in the manifest.
+### Baseline And Module Matrix
+### Divergence Register
+### Prompt, Schema, And LLM Review
+### Extraction And Canonicalization Review
+### Duplication Register
+### Contextual Architecture And Ownership Review
+### Candidate Decisions
+```
 
-### 1.2 Simplify NPC identity, normalization, and validation
+Create headings when their owning stage begins. Later stages may refine earlier
+sections but must not duplicate them. Use compact tables where they improve
+five-way comparison. Keep detailed evidence in the relevant review section and
+keep the divergence and duplication registers concise.
 
-- Keep the identity policy at `dnd.npcs.identity.v1`. Retain display
-  normalization, comparison keys, ID syntax, and deterministic ID derivation.
-  Validate nonblank canonical names, exact ID/name agreement, duplicate
-  canonical names, and duplicate IDs. Delete alias validation and alias-specific
-  issue locations/codes.
-- Keep NPC normalization at `dnd.npcs.normalize.v1`. Normalize the retained
-  display name, derive its ID, canonicalize and deduplicate source references,
-  consolidate records only by the canonical-name comparison key, preserve the
-  first stable record, and union exact evidence. Remove alias promotion,
-  relationship merge/rewrites, and their warning codes.
-- Keep the existing NPC shape and source-relatedness policy identifiers. The
-  shape validator checks only list presence, nonblank `id` and `name`, and
-  nonempty source references; source-relatedness grounds a record only through
-  its retained name. The existing source-ref validator remains the owner of
-  range validity.
-- Keep the default validation order domain-first, then JSON Schema, then the
-  advisory relatedness validator. Simplify only the validator implementations
-  and selections whose owned behavior changed.
+Each divergence-register entry must have:
 
-### 1.3 Separate durable registry provenance from model input
+- a stable identifier such as `D-01`;
+- affected modules;
+- observed difference;
+- provisional classification from the audit strategy;
+- evidence links;
+- impact or reason it may be harmless; and
+- status: open, confirmed, intentional, rejected, or superseded.
 
-- Continue to canonicalize and retain the complete redesigned NPC artifact for
-  registry validation, cache identity, manifests, and provenance.
-  `Registry.Digest()` remains the digest of that complete canonical artifact
-  when bound and remains empty when no registry was supplied.
-- Build a second structural JSON projection for model grounding with the exact
-  shape `{"npcs":[{"name":"Mira Thorn"}]}` in normalized registry order. It
-  contains names only: no IDs, source references, origin URI, aliases, or other
-  provenance. Generate it with typed values and `json.Marshal`, not string
-  concatenation. The empty projection is exactly `{"npcs":[]}`.
-- Add an explicit projection digest accessor. The digest is SHA-256 over the
-  exact projected bytes, including for an absent or empty registry, and
-  `PromptInput().Digest` must equal it. Keep full-artifact and projection digests
-  distinct even when their content happens to coincide.
-- Index `Registry.Lookup` by canonical name only. Retain immutable return values
-  and defensive copies. Remove alias indexing and alias-aware comments.
-- In spell extraction, combat extraction, and combat normalization, use the NPC
-  name-projection digest for the component-local checkpoint fingerprint because
-  it exactly describes the names that affect those operations. Keep the full
-  registry digest and count in manifest metadata for provenance. Framework-owned
-  generated-reference fingerprints may still invalidate a run when any upstream
-  artifact byte changes; do not broaden this stage into a framework cache
-  redesign.
+Each duplication-register entry must have:
 
-### 1.4 Update owners, consumers, and tests
+- a stable identifier such as `R-01`;
+- participating modules;
+- repeated responsibility;
+- exact, structural, or policy-duplication classification;
+- meaningful differences;
+- candidate owner, if any;
+- evidence links; and
+- provisional outcome: extract, harmonize without sharing, retain separately,
+  or defer.
 
-- Update NPC codec, schema, extractor, identity, validator, normalizer, registry,
-  registration, pipeline-integration, and CLI fixtures to the new shape. Adapt
-  spell and combat tests that construct `dnd.NPC` values so the repository
-  remains buildable, but do not change their own artifact contracts yet.
-- Add or rewrite focused tests for strict schema acceptance/rejection,
-  codec round trips, unchanged ID derivation for known names, name-only
-  consolidation, evidence union, registry immutability, and canonical-name-only
-  lookup. Rewrite the existing fixture for the minimal current contract; do not
-  retain the superseded rich fixture solely to test backwards incompatibility.
-- Test the projection as a data contract: it contains only ordered names,
-  equivalent normalized registries produce identical bytes and digest,
-  evidence/ID-only changes do not change the projection digest, and name/order
-  changes do. This is not authorization to snapshot assembled prompt prose or
-  prefix lengths.
-- Update `docs/integrations/dnd-npc-artifacts.md` to own the redesigned v1
-  durable schema. Update `docs/internal/modules.md` and
-  `docs/internal/llm.md` only for current NPC behavior that lands in this stage.
+Do not assign final severity until Stage 5 has traced ownership and evaluated
+impact.
 
-### Stage 1 completion criteria
+## Stage 1: Inventory And Convention Matrix
 
-- NPC durable and private schemas expose no removed enrichment fields.
-- The generated NPC prompt input contains names only while manifests retain
-  full registry provenance.
-- Spell and combat consumers accept generated or external registries in the
-  redesigned v1 shape; rich pre-redesign registry JSON fails strict decoding.
-- Focused NPC, spell-wiring, combat-wiring, integration, and CLI tests pass, and
-  `go test ./...` passes.
+### Objective
 
-## Stage 2: Cut Over Spell-Cast Extraction
+Establish the complete, factual five-module baseline and identify convention
+differences without yet recommending shared abstractions.
 
-### 2.1 Replace the spell contract and prompt
+### Required review
 
-- Reduce `dnd.SpellCast` to exactly `Caster`, `Spell`, and `SourceRefs`.
-- Rewrite `dnd_spells.v1.json` in place. Keep the durable codec at version v1,
-  name `notarius_dnd_spells_v1`, and its existing schema ID. Its strict record
-  contains only required `caster`, `spell`, and `source_refs`.
-- Rewrite `dnd_spells_llm.v1.json` with the same logical fields and model-facing
-  ranges. Keep the existing private key, ID, name, prompt version, and schema
-  path unchanged.
-- Reduce the private DTO, canonicalization, and mapping to the retained fields.
-  Continue assigning current source IDs, stable-ordering candidates by evidence,
-  and preserving semantically invalid candidates for deterministic validation.
-- Rewrite spell task and instruction assets around the narrow casting-event
-  boundary. Retain the spell catalog and name-only NPC projection as
-  disambiguation inputs, never evidence. Remove effect, outcome, and narrative
-  duties and remove the `common-dnd-immediate-resolution.md` message from the
-  spell manifest and spell asset registration. Do not delete the shared file in
-  this stage because combat still uses it.
+For every extractor, inspect:
 
-### 2.2 Simplify spell policies
+- production and test file inventory;
+- module key, artifact kind, capabilities, and execution class;
+- constructor inputs and retained prepared state;
+- option decoding and unknown-option behavior;
+- `ModuleSpec`, registration builder, and reference slots;
+- manifest metadata and checkpoint fingerprint providers;
+- exported versus package-private surface;
+- embedded asset registration;
+- principal extractor entry point and result type; and
+- package-local test organization.
 
-- Keep the spell shape policy identifier unchanged and validate only list
-  presence, nonblank caster/spell, and nonempty source references.
-- Keep catalog validation, source-reference validation, and source-relatedness
-  behavior and policy identities unchanged where their actual semantics are
-  already limited to spell name, caster, and evidence.
-- Simplify normalization and duplicate comparison to caster, catalog-canonical
-  spell name, and the complete valid evidence set. Remove all prose selection,
-  copying, fixture fields, and assertions. Preserve catalog fingerprints and
-  metadata.
-- Keep the exact inclusion rule from the feature roadmap: an actual casting or
-  unambiguous declared attempt is included; mentions, plans, rules discussion,
-  and catalog matches are not.
+Trace registration into production composition far enough to confirm that the
+declared contract is the one actually selected. Inspect neighboring packages
+only as required to verify an identity or ownership fact.
 
-### 2.3 Update owners and tests
+### Write to `audit.md`
 
-- Update spell codec, private schema, extractor, validators, normalizer, merge
-  and registration tests, pipeline integration, CLI output fixtures, and any
-  maintained examples to the minimal shape. Rewrite or delete rich-schema
-  fixtures rather than retaining them as compatibility cases.
-- Replace tests of effects and narrative descriptions with focused tests of the
-  retained contract: strict unknown-field rejection, codec round trips, current
-  source-ID assignment, catalog canonicalization, evidence ordering, duplicate
-  collapse, and NPC name-projection wiring.
-- Update `docs/integrations/dnd-spell-artifacts.md` as the canonical redesigned
-  v1 contract and update current internal module/LLM documentation for the
-  smaller prompt and response. Do not duplicate the spell catalog contract
-  owned by its existing integration document.
+Create `## Audit Results`, mark it `Status: In progress`, and add:
 
-### Stage 2 completion criteria
+1. `### Baseline And Module Matrix`
+   - one row per module;
+   - columns for every comparison dimension in the audit strategy;
+   - concise facts with links rather than judgments; and
+   - an explicit note where a dimension is not applicable.
+2. `### Divergence Register`
+   - record every observed organizational, naming, construction, registration,
+     provenance, or test-layout difference;
+   - classify only as required specialization, permitted variation, convention
+     drift, architectural divergence, or undetermined; and
+   - do not propose helper extraction in this stage.
 
-- No production spell type, schema, prompt, validator, normalizer, fixture, or
-  documentation contract refers to effect or narrative description.
-- Spell prompt/schema identities remain exactly their existing v1 values.
-- Focused spell and pipeline tests pass, and `go test ./...` passes.
-
-## Stage 3: Cut Over Combat-Turn Extraction
-
-### 3.1 Replace the combat contract and prompt
-
-- Reduce `dnd.CombatTurn` to exactly `Actor`, `TurnKind`, and `SourceRefs`.
-  Delete `CombatAction`, `CombatActionCategory`, their constants, and all
-  now-unused helpers. Retain the existing five `CombatTurnKind` values.
-- Rewrite `dnd_combat_turns.v1.json` in place. Keep the durable codec at version
-  v1, name `notarius_dnd_combat_turns_v1`, and its existing schema ID. Use a
-  strict required record with `actor`, `turn_kind`, and `source_refs` only.
-- Simplify the durable codec to direct strict encoding/decoding if its custom
-  wire representation exists only to distinguish removed nullable fields.
-  Preserve presence semantics for the top-level list and strict unknown-field
-  rejection.
-- Rewrite `dnd_combat_turns_llm.v1.json` while retaining its private schema and
-  prompt identities, and reduce the DTO/mapping to actor, turn kind, and
-  model-facing ranges. Keep the semantic mapping policy identifier unchanged.
-- Rewrite combat task/instruction assets around detecting ordered turns and
-  discrete interrupting events. Remove round, action, target, declaration,
-  resolution, outcome, and summary duties. Remove the shared immediate-
-  resolution message from the combat manifest.
-- Once both spell and combat manifests no longer use it, delete
-  `common-dnd-immediate-resolution.md` and its shared and module asset
-  registrations. Retain all other shared evidence, identity, reference, NPC,
-  and transcript assets in their cache-friendly order.
-
-### 3.2 Simplify combat normalization and validation
-
-- Keep combat normalization at `dnd.combat_turns.normalize.v1`. Continue to
-  display-normalize and registry-canonicalize actors, canonicalize evidence,
-  order records chronologically, and collapse exact duplicates by actor, turn
-  kind, and complete valid evidence. Delete action/target/prose normalization
-  and warning codes.
-- Keep the invariant policy identifier unchanged. It checks canonical actor
-  display, canonical and chronological evidence, stable event ordering, and
-  absence of duplicate event identities; it performs no nested-action checks.
-- Keep the shape policy identifier unchanged and validate only list presence,
-  nonblank actor, allowed turn kind, and nonempty source references.
-- Keep the source-relatedness policy identifier unchanged and compare only the
-  actor against cited transcript material. Remove declaration-token heuristics
-  and their now-unused helpers. Keep source-reference validation unchanged if
-  its semantics did not change.
-- Simplify merge/clone behavior to copy only retained values and source refs.
-  Preserve stable ordering and defensive ownership.
-
-### 3.3 Update owners and tests
-
-- Update codec, private schema, extractor, normalizer, validators, merge,
-  registration, pipeline integration, CLI fixtures, and examples to the minimal
-  shape. Rewrite or delete rich-schema fixtures rather than retaining them as
-  compatibility cases.
-- Delete tests whose sole policy was round/action/summary handling. Add or
-  rewrite focused tests for strict schemas, enum validation, mapping and
-  source-ID assignment, actor canonicalization through the redesigned NPC
-  registry, chronology, exact duplicate collapse, and invariant validation.
-- Update `docs/integrations/dnd-combat-turn-artifacts.md` as the canonical
-  redesigned v1 contract and update current internal module/LLM documentation
-  for the implemented behavior.
-
-### Stage 3 completion criteria
-
-- No production combat type, prompt, schema, policy, normalizer, fixture, or
-  current documentation refers to rounds, actions, summaries, declarations,
-  targets, resolutions, or action categories.
-- The immediate-resolution shared asset has no remaining registration or file.
-- Focused combat and pipeline tests pass, and `go test ./...` passes.
-
-## Stage 4: Complete The Repository-Wide Cutover
-
-### 4.1 Audit contract identity and stale surface area
-
-- Search code, embedded assets, tests, examples, and current documentation for
-  all removed field names and unintended v2 schema/prompt names. Removed fields
-  may remain only in historical ADR or roadmap context; active schema, prompt,
-  fixture, and current-behavior surfaces must describe the minimal v1 contract.
-- Verify the three artifact registrations still bind their original kinds and
-  exact redesigned Go types through extract, merge, normalize, codec, and
-  validators.
-- Verify all prompt manifests still point at their v1 private schemas, stable
-  reference material precedes module-variable material, and transcript content
-  remains last. Do not add a change-detector test for prompt message count,
-  prose, shared-prefix content, or prefix length.
-- Verify pre-redesign development checkpoints naturally miss through changed
-  prompt/schema content and generated-dependency fingerprints. Do not add
-  migration or compatibility handling for disposable pre-release state.
-
-### 4.2 Exercise representative assembled behavior
-
-- Keep unit case matrices at their owning schema, validator, normalizer, and
-  registry boundaries. At the assembled-pipeline boundary, retain only
-  representative tests proving each redesigned lane is registered, strict
-  JSON/schema failures are attributed to the correct validator, semantic
-  failures are attributed to the correct domain validator, and generated NPC
-  output is accepted by later spell/combat stages.
-- Cover an ordered multi-step run in which NPC extraction produces the minimal
-  artifact and spell/combat consume its name projection. Assert that downstream
-  event source references point only to the current transcript and not to NPC
-  registry evidence.
-- Verify CLI logical output and manifests retain the v1 schema/prompt identities,
-  full NPC registry provenance, and the relevant component fingerprints without
-  exposing prompt bodies or reference payloads.
-
-### 4.3 Finish documentation and lifecycle state
-
-- Reconcile the three integration documents, `docs/internal/modules.md`, and
-  `docs/internal/llm.md` with the final code. Correct any stale validator-order
-  descriptions while doing so; current docs must describe the registered order,
-  not preserve an older generic ordering example.
-- Mark ADR-0009 `Accepted` and the feature roadmap `Implemented` when the code,
-  tests, and current-behavior documentation all land. Remove the completed item
-  from `docs/roadmap/future.md`; retain the feature roadmap and ADR as design
-  rationale unless the repository's normal roadmap-retirement practice calls
-  for moving the completed roadmap later.
-- Record any human-reviewed rich/minimal model evaluation separately from CI
-  results. Evaluation may motivate later prompt tuning but does not reopen the
-  approved durable minimal field set within this implementation.
-
-### 4.4 Final verification
-
-Run, in order:
+Also add a short baseline-validation note recording:
 
 ```sh
-gofmt -w <changed Go files>
-go test ./...
+go test -count=1 ./...
 go vet ./...
 go build ./cmd/notarius
-go test -race ./internal/modules/dnd/...
+gofmt -l .
 git diff --check
 ```
 
-If the repository-wide race command exposes an unrelated, pre-existing failure,
-document it with the narrower affected package result; do not weaken or skip
-ordinary tests for the changed D&D packages.
+If a command fails, record the exact command, affected package, and concise
+failure classification. Do not fix it during the audit.
 
-### Stage 4 completion criteria
+### Completion gate
 
-- All three lanes use only their minimal v1 contracts from model response
-  through durable output.
-- No compatibility shim, removed-field policy, unintended v2 asset, or stale
-  current-behavior documentation remains.
-- Generated NPC references are name-only for LLM input and remain full-fidelity
-  for durable provenance.
-- Repository tests, vet, build, race checks for the changed domain, and diff
-  hygiene checks pass.
+Stage 1 is complete when every matrix cell is populated or marked not
+applicable, every observed baseline divergence has a stable register entry, and
+the repository's starting validation state is recorded.
+
+## Stage 2: Prompts, Schemas, And LLM Boundaries
+
+### Objective
+
+Determine whether the five modules consistently present stable context,
+references, lane instructions, transcripts, and structured-output contracts to
+the LLM, and identify exact or near-duplicate assets without changing them.
+
+### Required review
+
+For every extractor:
+
+- read the complete prompt manifest and every referenced local and shared
+  asset;
+- record the exact ordered message sequence, role, input, and cache-control
+  boundary;
+- compare shared message files by identity and bytes, not by paraphrased
+  meaning;
+- verify stable-to-variable ordering against `docs/internal/llm.md`;
+- map declared prompt inputs to module reference slots and generated inputs;
+- trace optional, required, empty, and generated reference projections;
+- inspect prompt and schema registration, hashing, and diagnostic redaction;
+- inspect the complete private response schema for identity, required fields,
+  nullability, strict objects, and semantic constraints;
+- map schema fields to private DTOs and response mapping; and
+- compare prompt and schema tests at their behavioral boundaries.
+
+Treat the scene-description whole-chunk evidence model as a specialization to
+explain, not a presumption of drift. Likewise, treat catalog and NPC registry
+inputs as subset-specific responsibilities unless evidence shows inconsistent
+handling of the same contract.
+
+### Write to `audit.md`
+
+Add `### Prompt, Schema, And LLM Review` containing:
+
+- a five-way prompt-order and cache-boundary table;
+- a prompt-input and reference-projection table;
+- a private-schema and DTO ownership table;
+- exact shared-asset usage;
+- local assets with identical or near-identical content;
+- content-safety and diagnostic observations; and
+- evidence-backed deviations from documented LLM conventions.
+
+Update the divergence register for confirmed or newly discovered prompt,
+schema, reference, provenance, and testing differences.
+
+Create `### Duplication Register` and add prompt-, schema-, metadata-, and
+asset-related candidates. For each prompt candidate, state whether the text is
+byte-identical, merely similar, or semantically different. Do not recommend a
+shared prompt asset unless all intended consumers should receive future edits
+atomically.
+
+### Completion gate
+
+Stage 2 is complete when every prompt message and input is accounted for, every
+private schema field has an owner, cache-prefix claims are based on exact
+message identity, and every asset-sharing candidate has a provisional keep or
+share outcome.
+
+## Stage 3: Extraction, Evidence, And Canonicalization
+
+### Objective
+
+Compare runtime extraction flow and identify repeated algorithms or policy
+without erasing artifact-specific semantics.
+
+### Required review
+
+For every extractor, inspect and trace:
+
+- request, context, source, chunk, and dependency validation;
+- construction-time state versus operation-time overrides;
+- structured request assembly and provider-error wrapping;
+- response-to-artifact mapping;
+- source identity attachment and unit-ID resolution;
+- evidence range validation assumptions;
+- canonical source-reference ordering and exact deduplication;
+- artifact ordering and deterministic tie-breakers;
+- enum or canonical-name handling;
+- warnings and diagnostics;
+- cloning and aliasing boundaries; and
+- focused extractor, model, canonicalization, and malformed-output tests.
+
+Compare complete algorithms rather than function names alone. For similar
+canonicalization helpers, identify which parts are:
+
+- identical source-reference mechanics;
+- artifact-specific mapping;
+- artifact-specific ordering policy; or
+- validation that belongs to a later validator rather than extraction.
+
+Use call traces and complexity data to inspect repeated scans, allocations,
+serialization, or high-cognitive-complexity paths. Report performance only when
+a plausible workload and complexity impact exist.
+
+### Write to `audit.md`
+
+Add `### Extraction And Canonicalization Review` containing:
+
+- a five-way extraction-flow table;
+- an evidence and ordering-policy table;
+- clone, mutation, error, and warning observations;
+- test-ownership comparisons; and
+- code-quality or performance candidates with concrete impact.
+
+Update the divergence register, revising earlier provisional classifications
+where runtime evidence explains or contradicts them.
+
+Expand the duplication register with request-validation, reference,
+canonicalization, ordering, deduplication, mapping, error, and test-support
+candidates. Give each candidate a provisional owner and explicitly identify the
+artifact-specific code that must remain local.
+
+### Completion gate
+
+Stage 3 is complete when the full extraction path of all five modules is
+accounted for, every similar canonicalization path has been decomposed into
+shared mechanics versus domain policy, and every runtime duplication candidate
+has a provisional ownership decision.
+
+## Stage 4: Contextual Architecture And Ownership Review
+
+### Objective
+
+Validate candidate findings against the surrounding D&D and framework
+architecture, and decide which apparent similarities should actually be shared.
+
+### Required review
+
+For every open divergence and duplication candidate:
+
+- trace callers and consumers;
+- inspect existing facilities in `internal/modules/dnd/shared`;
+- inspect focused D&D registry, identity, catalog, codec, normalize, and
+  validate packages as relevant;
+- verify the typed artifact and reference contracts;
+- inspect production registration, default validator composition, and
+  checkpoint fingerprint assembly;
+- verify documentation ownership and current durable contracts; and
+- check whether a proposed generic helper has a genuine domain-neutral owner.
+
+Apply the shared-code hierarchy from `audit.md`:
+
+1. module-owned artifact semantics;
+2. D&D-wide shared mechanics;
+3. focused subset-specific D&D packages; and
+4. framework-owned domain-neutral behavior.
+
+Reject or defer candidates whose API would require artifact-specific callbacks,
+type erasure, module-key branching, a broad configuration object, or speculative
+future consumers.
+
+### Write to `audit.md`
+
+Add `### Contextual Architecture And Ownership Review` containing:
+
+- traced ownership evidence for every open candidate;
+- dependency-direction and layer-boundary conclusions;
+- checkpoint, provenance, and reference compatibility conclusions;
+- documentation or test ownership implications; and
+- any product-contract questions that cannot be decided as refactors.
+
+Add `### Candidate Decisions`, with one row per divergence and duplication
+identifier. Choose exactly one outcome:
+
+- extract now;
+- harmonize without sharing;
+- retain intentionally separate;
+- reject as harmful abstraction; or
+- defer pending a named missing requirement or product decision.
+
+For extract or harmonize outcomes, specify:
+
+- target owner;
+- minimal responsibility and proposed API shape;
+- participating modules;
+- behavior that remains package-owned;
+- migration order;
+- relevant tests; and
+- principal risks.
+
+Update every register entry to confirmed, intentional, rejected, superseded, or
+explicitly deferred. No entry may remain merely open at the end of this stage.
+
+### Completion gate
+
+Stage 4 is complete when all candidates have traced ownership and a final
+keep/share/harmonize/defer decision, no proposed helper violates dependency
+direction, and all required product decisions are separated from executable
+refactoring recommendations.
+
+## Stage 5: Synthesis And Final Audit
+
+### Objective
+
+Turn the working evidence into one concise, internally consistent audit that
+can support roadmap decisions and a later implementation plan.
+
+### Required work
+
+1. Re-read the complete strategy and all working audit results.
+2. Recheck every cited file and symbol against the current working tree.
+3. Reconcile duplicate, overlapping, or contradictory observations.
+4. Assign severity only to confirmed actionable findings:
+   - **high:** correctness, security, data integrity, or architectural failure
+     with substantial impact;
+   - **medium:** meaningful drift, duplication, or design weakness likely to
+     cause defects or costly divergence;
+   - **low:** localized maintainability, clarity, test-quality, or
+     documentation issue with limited immediate impact.
+5. Keep optional improvements separate from findings.
+6. Confirm intentional specializations and rejected sharing candidates are
+   documented so future work does not repeatedly reopen them without evidence.
+7. Run the repository-wide validation commands from Stage 1 again and record
+   the final result. Do not modify code in response to failures.
+
+### Rewrite `audit.md`
+
+Preserve the strategy sections above `## Audit Results`, but replace the
+provisional working area with this final structure:
+
+```markdown
+## Audit Results
+
+Status: Complete
+
+### Executive Conclusion
+### Final Module Comparison Matrix
+### Prioritized Findings
+### Intentional Differences To Preserve
+### Shared-Code Decisions
+### Rejected Or Deferred Candidates
+### Recommended Refactoring Sequence
+### Validation And Residual Risks
+```
+
+Requirements for the final sections:
+
+- `Executive Conclusion` directly answers the user's three audit questions.
+- `Final Module Comparison Matrix` remains factual and compact.
+- `Prioritized Findings` follows the finding standard in `audit.md`, is ordered
+  by severity and impact, and cites exact evidence.
+- `Intentional Differences To Preserve` explains why harmonization would be
+  incorrect.
+- `Shared-Code Decisions` records extract and harmonize recommendations,
+  proposed ownership, and why sharing is better than continued separation.
+- `Rejected Or Deferred Candidates` records superficially attractive
+  abstractions and why they should not be pursued now.
+- `Recommended Refactoring Sequence` groups accepted recommendations into
+  independently safe, dependency-ordered scopes. It is a sequence, not a
+  decision-complete implementation plan.
+- `Validation And Residual Risks` records commands, results, limits of the
+  static audit, and any human or model-quality evaluation still needed.
+
+Remove the working divergence and duplication registers after their evidence
+has been incorporated into the final sections. Do not leave raw stage notes,
+provisional severities, or superseded conclusions in the completed audit.
+
+### Completion gate
+
+Stage 5 is complete when:
+
+- the final audit satisfies every deliverable and completion criterion in
+  `audit.md`;
+- the three user questions receive explicit answers;
+- every recommendation has evidence, ownership, scope, and rationale;
+- no finding relies only on similarity or stylistic preference;
+- validation results are current;
+- no code or current-behavior files changed; and
+- `git status --short` shows only the intended `audit.md` audit-result changes
+  plus any pre-existing user changes.
 
 ## Open Questions
 
-None. The feature roadmap, ADR, and fixed decisions above define the cutover,
-pre-release schema, evidence, projection, validation, testing, and documentation
-policies needed to implement each stage without further product decisions.
+None. The audit stages, working-document structure, decision rules, and final
+deliverable are fully specified.

docs/roadmap/minimal-dnd-extraction-contracts.md

@@ -1,220 +0,0 @@
-# Minimal D&D Extraction Contracts
-
-**Status:** Implemented
-
-## Intent
-
-Redesign the D&D spell, NPC, and combat-turn artifacts around the principle in
-[ADR-0009](../adr/0009-minimal-evidence-grounded-extraction-artifacts.md):
-each extractor should answer one narrow question with the smallest useful set
-of source-grounded fields.
-
-The redesign favors extraction precision, evidence quality, valid-output rate,
-smaller-model reliability, and lower prompt and response cost over descriptive
-richness. It removes synthesis responsibilities rather than preserving obsolete
-fields as optional, nullable, empty, or application-generated placeholders.
-
-## Goals
-
-- Make every model-produced field necessary to the artifact's core question.
-- Require direct transcript evidence for every extracted record.
-- Remove overlapping prose, inferred enrichment, and nested structures without
-  a demonstrated consumer.
-- Keep catalog and identity references as disambiguation aids rather than
-  evidence.
-- Preserve deterministic canonicalization, evidence ordering, exact
-  deduplication, identity assignment, and bounded domain diagnostics where
-  those responsibilities still apply.
-- Reduce downstream prompt material to the fields a consumer actually needs.
-- Keep the unpublished v1 identities while replacing their pre-release shapes
-  in place.
-
-## Non-Goals
-
-- Generating session narrative, rules analysis, biographies, relationship
-  graphs, encounter summaries, or prose descriptions.
-- Preserving removed fields for source compatibility through empty strings,
-  nullable values, or synthetic defaults.
-- Adding fuzzy entity resolution, LLM-assisted enrichment, or a general schema
-  migration framework.
-- Treating campaign references, catalogs, or earlier artifacts as evidence that
-  an event occurred in the current transcript.
-- Combining the three D&D artifact families into one model call.
-
-## Shared Contract Policy
-
-All three artifacts remain ordered lists. Each record contains at least one
-source reference, and the complete reference collection supports every
-model-produced field in that record. Source IDs continue to be assigned by the
-application from the current input; the model returns only source-unit ranges.
-
-Private LLM schemas remain strict about their transport envelope: required
-fields, JSON types, object and array shape, nullability where applicable, and
-unknown-field rejection. Deterministic validators continue to own semantic
-rules such as nonblank identities, catalog membership, enum membership,
-positive and resolvable source units, and canonical normalized invariants.
-
-Prompts retain the shared D&D evidence, identity, reference, NPC-grounding, and
-transcript assets that remain relevant. Module-specific task and instruction
-assets must delete duties associated with removed fields. The existing
-cache-friendly ordering keeps stable shared and reference material before the
-chunk-variable transcript.
-
-## Spell Cast
-
-The spell extractor answers:
-
-> Which spell was cast, by which in-world caster, and where is that event
-> established in the source?
-
-The artifact kind remains `dnd/spell-list`. Its durable schema remains v1.
-Each spell-cast record contains exactly:
-
-- `caster`: required nonblank in-world display identity;
-- `spell`: required nonblank canonical or catalog-resolvable spell name; and
-- `source_refs`: one or more current-source evidence ranges.
-
-The model-facing response contains the same fields except for application-owned
-`source_id` values within references. The current `effect` and
-`narrative_description` fields are removed from the private response, public Go
-type, durable schema, codec, validators, normalizer, fixtures, and integration
-contract.
-
-The inclusion boundary remains an actual casting event or an unambiguously
-declared casting attempt, not a spell mention, hypothetical plan, rules
-discussion, or catalog match. The spell catalog helps recognize and canonicalize
-the name but never establishes that a cast occurred.
-
-Normalization continues to canonicalize spell names, canonicalize evidence,
-and collapse exact duplicate events using caster, canonical spell name, and
-complete valid evidence. It performs no prose selection or merging.
-
-## NPC Registry
-
-The NPC extractor answers:
-
-> Which individually identifiable non-player characters are established in the
-> source, and where is each identity established?
-
-The artifact kind remains `dnd/npc-list`. Its durable schema remains v1.
-Each durable NPC record contains exactly:
-
-- `id`: deterministic application-assigned identity derived under the NPC
-  identity policy;
-- `name`: required nonblank source-supported display identity; and
-- `source_refs`: one or more evidence ranges supporting that identity.
-
-The private model response omits `id` and reference `source_id` values. A
-`name` may be a proper name or a stable, individually distinguishing title or
-alias supported by the transcript. The extractor does not invent descriptive
-labels for anonymous creatures, crowds, or generic roles.
-
-The current `aliases`, `description`, and `relationships` fields are removed
-from the private response, public Go type, durable schema, codec, validators,
-normalizer, registry, fixtures, and integration contract. Normalization
-consolidates only identities supported by the retained name policy and unions
-exact evidence; it does not infer alias equivalence or relationships.
-
-Spell and combat consumers receive a prompt projection containing only the
-canonical NPC names needed for identity grounding. Application-owned NPC IDs
-remain available to deterministic registry and normalization code but are not
-sent to a model that cannot return or otherwise consume them. NPC source
-references remain provenance in the durable registry and are not included as
-current-transcript evidence or copied into downstream event artifacts.
-
-Encounter context is deliberately not a scalar NPC registry field. Dialogue,
-combat alignment, presence, or third-party mention can vary across occurrences.
-If a demonstrated consumer needs that information, add a separate ordered
-NPC-occurrence artifact whose records contain `name`, a small mutually
-exclusive context enum, and `source_refs`. A candidate starting vocabulary is
-`dialogue`, `combat_ally`, `combat_opponent`, `noncombat_presence`, `mentioned`,
-and `other`; its exact semantics require a separate feature decision.
-
-## Combat Event
-
-The combat extractor answers:
-
-> Which in-world participant took a turn or discrete interrupting combat
-> event, what kind of event was it, and where is it established in the source?
-
-The existing `dnd/combat-turn-list` artifact kind and v1 durable-schema identity
-remain. Each record contains exactly:
-
-- `actor`: required nonblank in-world display identity;
-- `turn_kind`: one of `turn`, `reaction`, `legendary_action`, `lair_action`, or
-  `other`; and
-- `source_refs`: one or more current-source evidence ranges.
-
-The current `round`, `actions`, and `summary` fields, including nested action
-categories, declarations, targets, and resolutions, are removed from the
-private response, public Go types, durable schema, codec, validators,
-normalizer, fixtures, and integration contract.
-
-Normalization continues to display-normalize and registry-canonicalize actors,
-canonicalize evidence, order events by source position, and collapse exact
-duplicates using actor, turn kind, and complete valid evidence. It no longer
-normalizes targets, declarations, summaries, or resolutions.
-
-If action-level facts later have a demonstrated consumer, they belong in a
-separate combat-action artifact rather than restoring a nested synthesis
-contract to combat-turn detection. Spell casts and future item events remain
-owned by their dedicated artifact lanes.
-
-## Pre-Release Schema Policy
-
-Notarius and these contracts are pre-release. The existing v1 artifacts and
-private model-response schemas have not been published as compatibility
-contracts, so their shapes change in place. The implementation does not retain
-the rich pre-redesign shape, add v2 assets, migrate old output, support multiple
-versions, or preserve old fixtures solely for compatibility testing.
-
-Artifact kinds, schema keys and IDs, schema names and versions, prompt IDs and
-versions, module keys, capabilities, and media types all remain unchanged.
-Changing prompt and schema content invalidates the relevant content-addressed
-development state; any remaining pre-redesign local output or cache is
-disposable and may be regenerated.
-
-| Lane | Durable v1 schema | Private model-response v1 schema | Prompt ID |
-| --- | --- | --- | --- |
-| Spell cast | ID `notarius.dnd.spells`, name `notarius_dnd_spells_v1` | key `dnd_spells`, ID `notarius.dnd.spells`, name `notarius_dnd_spells_v1` | `dnd.spells` |
-| NPC registry | ID `notarius.dnd.npcs`, name `notarius_dnd_npcs_v1` | key `dnd_npcs_llm`, ID `notarius.dnd.npcs.llm`, name `notarius_dnd_npcs_llm_v1` | `dnd.npcs` |
-| Combat event | ID `notarius.dnd.combat_turns`, name `notarius_dnd_combat_turns_v1` | key `dnd_combat_turns_llm`, ID `notarius.dnd.combat_turns.llm`, name `notarius_dnd_combat_turns_llm_v1` | `dnd.combat_turns` |
-
-## Quality And Evaluation
-
-The implemented contract should protect the remaining meaningful risks:
-
-- private schemas accept only the new structural envelopes;
-- durable codecs strictly round-trip the redesigned contracts;
-- domain validators own blank identities, enum and catalog membership, and
-  invalid evidence;
-- normalizers preserve source-grounded values while applying only their stated
-  deterministic transformations;
-- generated NPC references expose the minimal identity projection and never
-  become event evidence; and
-- representative assembled pipelines attribute retries and rejections to the
-  owning domain boundary.
-
-Post-cutover model evaluation should compare the rich-schema baseline and the
-minimal-schema result on a small human-reviewed transcript set using:
-
-- event/entity precision and recall;
-- caster, NPC, and combat-actor attribution accuracy;
-- source-range validity and evidence sufficiency;
-- catalog and enum accuracy;
-- structurally valid completion rate and exhausted-retry rate;
-- unsupported-claim rate; and
-- input/output tokens, latency, and model cost.
-
-The evaluation exists to identify prompt or model-quality follow-up work, not
-to gate the approved minimal contract on live-provider behavior. Human review
-and live model calls are evaluation aids rather than deterministic CI gates.
-
-## Documentation Ownership
-
-The spell, NPC, and combat integration documents are the canonical owners of
-their redesigned durable schemas. Internal LLM and module documents own the
-corresponding current prompt, DTO, validator, normalizer, and NPC
-prompt-projection behavior. This roadmap records the implemented design and
-policy; [the implementation plan](implementation.md) records the completed
-sequencing and completion criteria.

examples/dnd-npc-interactions.config.yml

@@ -0,0 +1,34 @@
+version: 3
+output:
+  directory: ./notarius-output
+cache:
+  chunk_plans:
+    mode: bypass
+  checkpoints:
+    enabled: false
+    directory: ""
+debug:
+  directory: ./notarius-debug
+pipelines:
+  dnd-npc-interactions:
+    input: seriatim
+    steps:
+      - id: identify-npcs
+        artifacts:
+          npcs:
+            extract:
+              module: dnd/npcs
+              retries: 2
+            normalize: dnd/npcs
+      - id: extract-interactions
+        references:
+          npcs:
+            artifact:
+              step: identify-npcs
+              lane: npcs
+        artifacts:
+          interactions:
+            extract:
+              module: dnd/npc-interactions
+              retries: 2
+            normalize: dnd/npc-interactions

examples/dnd-scene-chunk-map.config.yml

@@ -0,0 +1,23 @@
+version: 3
+output:
+  directory: ./notarius-output
+cache:
+  chunk_plans:
+    mode: bypass
+  checkpoints:
+    enabled: false
+    directory: ""
+debug:
+  directory: ./notarius-debug
+pipelines:
+  dnd-scene-chunk-map:
+    input: seriatim
+    chunk: dnd/scenes
+    output:
+      module: json
+      options:
+        include_chunk_map: true
+    artifacts:
+      spells:
+        extract: dnd/spells
+        normalize: dnd/spells

examples/dnd-scene-descriptions.config.yml

@@ -0,0 +1,20 @@
+version: 3
+output:
+  directory: ./notarius-output
+cache:
+  chunk_plans:
+    mode: bypass
+    directory: ./notarius-cache/chunk-plans
+  checkpoints:
+    enabled: false
+    directory: ./notarius-cache/checkpoints
+debug:
+  directory: ./notarius-debug
+pipelines:
+  dnd-scene-descriptions:
+    input: seriatim
+    chunk: dnd/scenes
+    artifacts:
+      scene-descriptions:
+        extract: dnd/scene-descriptions
+        normalize: dnd/scene-descriptions

internal/cli/dnd_interactions_contract_test.go

@@ -0,0 +1,135 @@
+package cli
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	interactioncodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcinteractions"
+	interactionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcinteractions"
+	npcextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcs"
+	interactionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcinteractions"
+	npcnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcs"
+)
+
+func TestProductionNPCInteractionPipelineResolvesAndPrepares(t *testing.T) {
+	components := productionTestComponents(t)
+	resolved, err := pipeline.ResolvePipeline(npcInteractionProfile(pipeline.GeneratedReference("npcs", "npcs")), pipeline.ResolveOptions{}, catalogFromRegistries(components.registries))
+	if err != nil {
+		t.Fatalf("ResolvePipeline() error = %v", err)
+	}
+	if len(resolved.Steps) != 2 || len(resolved.Steps[1].ArtifactLanes) != 1 {
+		t.Fatalf("resolved pipeline = %#v", resolved)
+	}
+	lane := resolved.Steps[1].ArtifactLanes[0]
+	if lane.ArtifactKind != dnd.NPCInteractionListKind || lane.Extract.Module != interactionextract.Key || lane.Normalize.Module != interactionnormalize.Key {
+		t.Fatalf("interaction lane = %#v", lane)
+	}
+	for _, bindings := range [][]pipeline.ReferenceBinding{lane.ExtractReferences.Bindings, lane.NormalizeReferences.Bindings} {
+		if len(bindings) != 1 || bindings[0].SlotName != "npcs" || bindings[0].Artifact == nil || bindings[0].Artifact.Step != "npcs" || bindings[0].Artifact.Lane != "npcs" {
+			t.Fatalf("generated bindings = %#v", bindings)
+		}
+	}
+	if _, err := pipeline.Prepare(resolved, components.registries, pipeline.ModuleDependencies{LLM: &productionFakeLLMClient{}}); err != nil {
+		t.Fatalf("Prepare() error = %v", err)
+	}
+
+	catalog := catalogFromRegistries(components.registries)
+	codecSpec, ok := catalog.ArtifactCodecs.Spec(dnd.NPCInteractionListKind)
+	if !ok || codecSpec.Schema.ID != interactioncodec.SchemaID || codecSpec.Schema.Version != interactioncodec.SchemaVersion {
+		t.Fatalf("NPC interaction codec spec = %#v", codecSpec)
+	}
+}
+
+func TestProductionNPCInteractionReferencesRequireEarlierCompatibleProducer(t *testing.T) {
+	components := productionTestComponents(t)
+	catalog := catalogFromRegistries(components.registries)
+	laterProfile := npcInteractionProfile(pipeline.GeneratedReference("npcs", "npcs"))
+	laterProfile.Steps[0].ID = "seed"
+	laterProfile.Steps[0].Artifacts["seed"] = laterProfile.Steps[0].Artifacts["npcs"]
+	delete(laterProfile.Steps[0].Artifacts, "npcs")
+	laterProfile.Steps = append(laterProfile.Steps, pipeline.PipelineStepProfile{ID: "future", Artifacts: map[string]pipeline.ArtifactLaneProfile{
+		"npcs": {Extract: pipeline.Binding(npcextract.Key), Normalize: pipeline.Binding(npcnormalize.Key)},
+	}})
+	laterProfile.Steps[1].References["npcs"] = pipeline.GeneratedReference("future", "npcs")
+	tests := []struct {
+		name    string
+		profile pipeline.PipelineProfile
+		want    string
+	}{
+		{name: "missing", profile: npcInteractionProfile(pipeline.ReferenceSource{}), want: "source must not be empty"},
+		{name: "same step", profile: npcInteractionProfile(pipeline.GeneratedReference("interactions", "interactions")), want: "earlier step"},
+		{name: "later step", profile: laterProfile, want: "earlier step"},
+		{name: "wrong artifact kind", profile: npcInteractionProfile(pipeline.GeneratedReference("npcs", "npcs")), want: "does not accept artifact kind"},
+	}
+	tests[3].profile.Steps[0].Artifacts["npcs"] = pipeline.ArtifactLaneProfile{Extract: pipeline.Binding("dnd/spells")}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			_, err := pipeline.ResolvePipeline(test.profile, pipeline.ResolveOptions{}, catalog)
+			if err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("ResolvePipeline() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+}
+
+func TestProductionNPCInteractionReferencesRejectIncompatibleExternalRegistries(t *testing.T) {
+	components := productionTestComponents(t)
+	catalog := catalogFromRegistries(components.registries)
+	root := t.TempDir()
+	for _, test := range []struct {
+		name    string
+		file    string
+		content string
+		prepare bool
+		want    string
+	}{
+		{name: "media type", file: "registry.txt", content: "not JSON", want: "media type"},
+		{name: "artifact schema", file: "registry.json", content: `{"npcs":[{"name":"missing required fields"}]}`, prepare: true, want: "NPC registry"},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			path := filepath.Join(root, test.file)
+			if err := os.WriteFile(path, []byte(test.content), 0o600); err != nil {
+				t.Fatal(err)
+			}
+			resolved, err := pipeline.ResolvePipeline(npcInteractionProfile(pipeline.ExternalReference(path)), pipeline.ResolveOptions{}, catalog)
+			if err != nil {
+				t.Fatalf("ResolvePipeline() error = %v", err)
+			}
+			materialized, _, err := pipeline.MaterializeReferences(resolved, catalog, pipeline.ReferenceMaterializationOptions{})
+			if !test.prepare {
+				if err == nil || !strings.Contains(err.Error(), test.want) {
+					t.Fatalf("MaterializeReferences() error = %v, want %q", err, test.want)
+				}
+				return
+			}
+			if err != nil {
+				t.Fatalf("MaterializeReferences() error = %v", err)
+			}
+			if _, err := pipeline.Prepare(materialized, components.registries, pipeline.ModuleDependencies{LLM: &productionFakeLLMClient{}}); err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("Prepare() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+}
+
+func npcInteractionProfile(reference pipeline.ReferenceSource) pipeline.PipelineProfile {
+	profile := pipeline.PipelineProfile{
+		ID:     "dnd-npc-interactions",
+		Input:  pipeline.Binding("seriatim"),
+		Chunk:  pipeline.ModuleBinding{Module: "generic", Options: map[string]any{"max_units": 1}},
+		Output: pipeline.Binding("json"),
+		Steps: []pipeline.PipelineStepProfile{
+			{ID: "npcs", Artifacts: map[string]pipeline.ArtifactLaneProfile{
+				"npcs": {Extract: pipeline.Binding(npcextract.Key), Normalize: pipeline.Binding(npcnormalize.Key)},
+			}},
+			{ID: "interactions", References: map[string]pipeline.ReferenceSource{"npcs": reference}, Artifacts: map[string]pipeline.ArtifactLaneProfile{
+				"interactions": {Extract: pipeline.Binding(interactionextract.Key), Normalize: pipeline.Binding(interactionnormalize.Key)},
+			}},
+		},
+	}
+	return profile
+}

internal/cli/dnd_scene_descriptions_contract_test.go

@@ -0,0 +1,111 @@
+package cli
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/config"
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	scenecodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/scenedescriptions"
+	sceneextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/scenedescriptions"
+	scenenormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/scenedescriptions"
+)
+
+func TestProductionSceneDescriptionWorkflow(t *testing.T) {
+	components := productionTestComponents(t)
+	cfg := config.Default()
+	cfg.Pipelines["scene-descriptions"] = pipeline.PipelineProfile{
+		ID:     "scene-descriptions",
+		Input:  pipeline.Binding("seriatim"),
+		Chunk:  pipeline.ModuleBinding{Module: "generic", Options: map[string]any{"max_units": 1}},
+		Output: pipeline.Binding("json"),
+		Artifacts: map[string]pipeline.ArtifactLaneProfile{
+			"scene-descriptions": {
+				Extract:   pipeline.Binding(sceneextract.Key),
+				Normalize: pipeline.Binding(scenenormalize.Key),
+			},
+		},
+	}
+	effective, err := cfg.Resolve(config.ResolveInput{PipelineID: "scene-descriptions", Catalog: catalogFromRegistries(components.registries)})
+	if err != nil {
+		t.Fatalf("Resolve() error = %v", err)
+	}
+	lane := effective.ResolvedPipeline.Steps[0].ArtifactLanes[0]
+	if lane.ArtifactKind != dnd.SceneDescriptionListKind || lane.Extract.Module != sceneextract.Key || lane.Merge.Module != pipeline.DefaultMergeModule || lane.Normalize.Module != scenenormalize.Key {
+		t.Fatalf("resolved lane = %#v, want production scene-description composition", lane)
+	}
+	if len(lane.ExtractReferences.Bindings) != 0 || len(lane.NormalizeReferences.Bindings) != 0 {
+		t.Fatalf("resolved references = %#v / %#v, want no generated or required references", lane.ExtractReferences, lane.NormalizeReferences)
+	}
+
+	prepared, err := pipeline.Prepare(effective.ResolvedPipeline, components.registries, pipeline.ModuleDependencies{LLM: sceneDescriptionLLM{}})
+	if err != nil {
+		t.Fatalf("Prepare() error = %v", err)
+	}
+	output, err := pipeline.New().Run(context.Background(), pipeline.RunInput{
+		Prepared:       prepared,
+		RawInput:       readRepositoryFile(t, "examples", "seriatim-minimal-transcript.json"),
+		ChunkCacheMode: pipeline.ChunkCacheBypass,
+	})
+	if err != nil {
+		t.Fatalf("Run() error = %v", err)
+	}
+	if output.Manifest.ValidationStatus != "approved" || len(output.Rejected) != 0 || len(output.NormalizeOutputs) != 1 {
+		t.Fatalf("run output = %#v, want one approved normalized artifact", output)
+	}
+	normalizedOutput := output.NormalizeOutputs[0]
+	if normalizedOutput.NormalizerKey != scenenormalize.Key || normalizedOutput.Artifact.Kind != dnd.SceneDescriptionListKind || normalizedOutput.Artifact.Schema.ID != scenecodec.SchemaID || normalizedOutput.Artifact.Schema.Name != scenecodec.SchemaName || normalizedOutput.Artifact.Schema.Version != scenecodec.SchemaVersion {
+		t.Fatalf("normalized output = %#v, want registered durable scene-description schema", normalizedOutput)
+	}
+
+	var value dnd.SceneDescriptionList
+	if err := json.Unmarshal(normalizedOutput.Artifact.Content, &value); err != nil {
+		t.Fatalf("decode normalized artifact: %v", err)
+	}
+	want := dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{
+		{ID: "chunk-000001", SourceRef: source.SourceRef{SourceID: "session-alpha", StartUnitID: 1, EndUnitID: 1}, Kind: dnd.SceneKindNarrative, Title: "Aria casts Cure Wounds", Summary: "Aria casts Cure Wounds."},
+		{ID: "chunk-000002", SourceRef: source.SourceRef{SourceID: "session-alpha", StartUnitID: 2, EndUnitID: 2}, Kind: dnd.SceneKindCombat, Title: "Bandit mage casts Shield", Summary: "The bandit mage casts Shield."},
+	}}
+	if !reflect.DeepEqual(value, want) {
+		t.Fatalf("normalized scene descriptions = %#v, want %#v", value, want)
+	}
+	durable := decodeAssembledOutput[dnd.SceneDescriptionList](t, output.OutputFiles, "lanes/scene-descriptions.json")
+	if !reflect.DeepEqual(durable, want) {
+		t.Fatalf("durable output payload = %#v, want %#v", durable, want)
+	}
+	if len(output.Warnings) != 0 {
+		t.Fatalf("warnings = %#v, want grounded descriptions without warnings", output.Warnings)
+	}
+}
+
+type sceneDescriptionLLM struct{}
+
+func (sceneDescriptionLLM) CompleteStructured(ctx context.Context, req contracts.StructuredCompletionRequest, out any) (contracts.StructuredCompletionResponse, error) {
+	if err := ctx.Err(); err != nil {
+		return contracts.StructuredCompletionResponse{}, err
+	}
+	if req.PromptID != sceneextract.PromptID {
+		return contracts.StructuredCompletionResponse{}, fmt.Errorf("unexpected prompt %q", req.PromptID)
+	}
+	transcript := string(req.Inputs["transcript"].Content)
+	var content string
+	switch {
+	case strings.Contains(transcript, "Cure Wounds"):
+		content = `{"kind":"narrative","title":"  Aria casts Cure Wounds  ","summary":"  Aria casts Cure Wounds.  "}`
+	case strings.Contains(transcript, "Shield"):
+		content = `{"kind":"combat","title":"Bandit mage casts Shield","summary":"The bandit mage casts Shield."}`
+	default:
+		return contracts.StructuredCompletionResponse{}, fmt.Errorf("unexpected transcript material %q", transcript)
+	}
+	if err := json.Unmarshal([]byte(content), out); err != nil {
+		return contracts.StructuredCompletionResponse{}, fmt.Errorf("populate structured response: %w", err)
+	}
+	return contracts.StructuredCompletionResponse{Content: []byte(content), Provider: "test", Model: "deterministic", ProfileID: req.ProfileID}, nil
+}

internal/cli/production_contract_test.go

@@ -18,6 +18,7 @@ import (
 
 	"gitea.maximumdirect.net/eric/notarius/internal/core/artifacts"
 	"gitea.maximumdirect.net/eric/notarius/internal/core/config"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/chunkmap"
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
@@ -457,7 +458,7 @@ func TestProductionNormalizeValidatorOverrideRemainsAuthoritative(t *testing.T)
 	t.Fatalf("resolved validator chains = %#v, want normalize chain for %q", effective.ResolvedPipeline.ValidatorChains, spellnormalize.Key)
 }
 
-func TestProductionSceneRunRecordsChunkerWarningsAndProvenance(t *testing.T) {
+func TestProductionSceneRunRecordsAnnotationFreeChunkPlanAndProvenance(t *testing.T) {
 	outputRoot := filepath.Join(t.TempDir(), "output")
 	configPath := writeProductionContractConfig(t, productionRunConfig(outputRoot, "dnd/scenes"))
 	fake := &productionFakeLLMClient{}
@@ -481,11 +482,40 @@ func TestProductionSceneRunRecordsChunkerWarningsAndProvenance(t *testing.T) {
 	if got := manifest.ChunkPlan.ProducerMetadata["response_schema_id"]; got != scenes.ResponseSchemaID {
 		t.Fatalf("chunk producer schema metadata = %#v, want %q", got, scenes.ResponseSchemaID)
 	}
+	index := readProductionJSON[productionChunkMapIndex](t, filepath.Join(outputRoot, productionRunID, "index.json"))
+	if index.ChunkMap == nil || index.ChunkMap.ArtifactKind != chunkmap.ArtifactKind || index.ChunkMap.File != "chunk-map.json" || index.ChunkMap.MediaType != chunkmap.MediaType || index.ChunkMap.SchemaID != chunkmap.SchemaID || index.ChunkMap.SchemaName != chunkmap.SchemaName || index.ChunkMap.SchemaVersion != chunkmap.SchemaVersion {
+		t.Fatalf("chunk map index = %#v, want fixed chunk map descriptor", index.ChunkMap)
+	}
+	for _, output := range index.OutputFiles {
+		if output.File == index.ChunkMap.File {
+			t.Fatalf("lane output files = %#v, want no chunk map", index.OutputFiles)
+		}
+	}
+	content, err := os.ReadFile(filepath.Join(outputRoot, productionRunID, index.ChunkMap.File))
+	if err != nil {
+		t.Fatal(err)
+	}
+	chunkMap, err := chunkmap.New().Decode(content)
+	if err != nil {
+		t.Fatalf("Decode(chunk map) error = %v", err)
+	}
+	if chunkMap.SourceID != "session-alpha" || chunkMap.SourceDigest != manifest.ChunkPlan.SourceDigest || chunkMap.PlanDigest != manifest.ChunkPlan.PlanDigest || chunkMap.RequestedChunker != scenes.Key || chunkMap.Producer.InputModule != "seriatim" || chunkMap.Producer.ChunkModule != scenes.Key || chunkMap.Producer.LLMProfile != manifest.ChunkPlan.ProducerLLMProfile {
+		t.Fatalf("chunk map identity and producer = %#v, want accepted scene plan provenance", chunkMap)
+	}
+	if len(chunkMap.Chunks) != 1 || chunkMap.Chunks[0].ID != "chunk-000001" || chunkMap.Chunks[0].Index != 0 || chunkMap.Chunks[0].SourceRef.SourceID != "session-alpha" || chunkMap.Chunks[0].SourceRef.StartUnitID != 1 || chunkMap.Chunks[0].SourceRef.EndUnitID != 2 || chunkMap.Chunks[0].UnitCount != 2 {
+		t.Fatalf("chunk map chunks = %#v, want one stable accepted scene range", chunkMap.Chunks)
+	}
+	if len(chunkMap.PlanAnnotations) != 0 {
+		t.Fatalf("chunk map plan annotations = %#v, want none", chunkMap.PlanAnnotations)
+	}
+	if len(chunkMap.Chunks[0].Annotations) != 0 {
+		t.Fatalf("chunk map range annotations = %#v, want none", chunkMap.Chunks[0].Annotations)
+	}
 	warnings := readProductionJSON[struct {
 		Warnings []contracts.Warning `json:"warnings"`
 	}](t, filepath.Join(outputRoot, productionRunID, "warnings.json"))
-	if len(warnings.Warnings) != 1 || warnings.Warnings[0].ReasonCode != "scene_boundary_caveat" {
-		t.Fatalf("warnings = %#v, want one scene boundary warning", warnings.Warnings)
+	if len(warnings.Warnings) != 0 {
+		t.Fatalf("warnings = %#v, want none", warnings.Warnings)
 	}
 	if len(fake.requestsFor(scenes.PromptID)) != 1 || len(fake.requestsFor(spells.PromptID)) != 1 {
 		t.Fatalf("fake prompt requests = %#v, want one scene and one spell request", fake.requestPrompts())
@@ -506,6 +536,9 @@ func maintainedExampleFiles(t *testing.T) []maintainedExample {
 		{name: "npcs", path: repositoryPath("examples", "dnd-npcs.config.yml"), pipelineIDs: []string{"dnd-session"}},
 		{name: "combat", path: repositoryPath("examples", "dnd-combat-turns.config.yml"), pipelineIDs: []string{"dnd-combat"}},
 		{name: "npc-grounded", path: repositoryPath("examples", "dnd-npc-grounded.config.yml"), pipelineIDs: []string{"dnd-npc-grounded"}},
+		{name: "npc-interactions", path: repositoryPath("examples", "dnd-npc-interactions.config.yml"), pipelineIDs: []string{"dnd-npc-interactions"}},
+		{name: "scene-descriptions", path: repositoryPath("examples", "dnd-scene-descriptions.config.yml"), pipelineIDs: []string{"dnd-scene-descriptions"}},
+		{name: "scene-chunk-map", path: repositoryPath("examples", "dnd-scene-chunk-map.config.yml"), pipelineIDs: []string{"dnd-scene-chunk-map"}},
 	}
 }
 
@@ -587,12 +620,30 @@ pipelines:
   dnd-session:
     input: seriatim
     chunk: %s
+    output:
+      module: json
+      options:
+        include_chunk_map: true
     artifacts:
       spells:
         extract: dnd/spells
 `, outputRoot, filepath.Join(filepath.Dir(outputRoot), "debug"), chunkModule)
 }
 
+type productionChunkMapIndex struct {
+	OutputFiles []struct {
+		File string `json:"file"`
+	} `json:"output_files"`
+	ChunkMap *struct {
+		ArtifactKind  contracts.ArtifactKind `json:"artifact_kind"`
+		File          string                 `json:"file"`
+		MediaType     string                 `json:"media_type"`
+		SchemaID      string                 `json:"schema_id"`
+		SchemaName    string                 `json:"schema_name"`
+		SchemaVersion string                 `json:"schema_version"`
+	} `json:"chunk_map"`
+}
+
 func writeProductionContractConfig(t *testing.T, content string) string {
 	t.Helper()
 	path := filepath.Join(t.TempDir(), "config.yml")
@@ -667,7 +718,7 @@ func (client *productionFakeLLMClient) CompleteStructured(ctx context.Context, r
 	var content []byte
 	switch req.PromptID {
 	case scenes.PromptID:
-		content = []byte(`{"scenes":[{"start_unit_id":1,"end_unit_id":2,"short_title":"Opening scene","primary_mode":"Narrative","main_participants":["Aria"],"summary":"The session opens.","boundary_note":"The opening covers the available transcript.","boundary_confidence":"High"}],"boundary_caveats":["The opening boundary is inferred from the short transcript."]}`)
+		content = []byte(`{"scenes":[{"start_unit_id":1,"end_unit_id":2}]}`)
 	case spells.PromptID:
 		if client.spellResponse != "" {
 			content = []byte(client.spellResponse)

internal/cli/state_hardening_test.go

@@ -24,7 +24,7 @@ import (
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
 )
 
-const stateTestDigest = "sha256:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"
+const stateTestDigest = "sha256:e511d8906649b78eb639b11215fa57a9652a1a64f4aefa3ed68320dbda46f439"
 
 func TestRunStateSurfaceMatrix(t *testing.T) {
 	for _, debug := range []bool{false, true} {
@@ -857,7 +857,13 @@ type stateTestInput struct{}
 
 func (stateTestInput) Key() string { return "test/input" }
 func (stateTestInput) Parse(_ context.Context, req contracts.ParseRequest) (*source.SourceDocument, error) {
-	return &source.SourceDocument{ID: "source", Kind: "text", Format: "text/plain", Digest: stateTestDigest, Units: []source.SourceUnit{{ID: 1, Kind: "text", Text: string(req.Raw), Ref: source.SourceRef{SourceID: "source", StartUnitID: 1, EndUnitID: 1}}}}, nil
+	doc := &source.SourceDocument{ID: "source", Kind: "text", Format: "text/plain", Units: []source.SourceUnit{{ID: 1, Kind: "text", Text: string(req.Raw), Ref: source.SourceRef{SourceID: "source", StartUnitID: 1, EndUnitID: 1}}}}
+	digest, err := source.DigestDocument(doc)
+	if err != nil {
+		return nil, err
+	}
+	doc.Digest = digest
+	return doc, nil
 }
 
 type stateTestChunker struct{ harness *stateTestHarness }

internal/core/artifacts/artifacts_test.go

@@ -20,7 +20,7 @@ func TestRunManifestOmitsEmptyOptionalFields(t *testing.T) {
 func TestRunManifestChunkPlanIsAdditiveAndOmitsPlanContent(t *testing.T) {
 	manifest := RunManifest{ChunkPlan: &ChunkPlanManifest{
 		Mode: "auto", Action: "reused", SourceDigest: "sha256:source", PlanDigest: "sha256:plan",
-		PlanSchemaVersion: "notarius.chunk-plan.v1", RequestedModule: "chunk/current",
+		PlanSchemaVersion: "notarius.chunk-plan.v2", RequestedModule: "chunk/current",
 		ProducerInputModule: "input/original", ProducerModule: "chunk/original",
 	}}
 	encoded, err := json.Marshal(manifest)

internal/framework/chunkmap/assets/schemas/source_chunk_map.v1.json

@@ -0,0 +1,65 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "notarius.source.chunk_map",
+  "title": "notarius_source_chunk_map_v1",
+  "type": "object",
+  "additionalProperties": false,
+  "required": [
+    "source_id",
+    "source_digest",
+    "plan_digest",
+    "requested_chunker",
+    "producer",
+    "plan_annotations",
+    "chunks"
+  ],
+  "properties": {
+    "source_id": {"type": "string", "minLength": 1},
+    "source_digest": {"type": "string", "pattern": "^sha256:[0-9a-f]{64}$"},
+    "plan_digest": {"type": "string", "pattern": "^sha256:[0-9a-f]{64}$"},
+    "requested_chunker": {"type": "string", "minLength": 1},
+    "producer": {
+      "type": "object",
+      "additionalProperties": false,
+      "required": ["input_module", "chunk_module"],
+      "properties": {
+        "input_module": {"type": "string", "minLength": 1},
+        "chunk_module": {"type": "string", "minLength": 1},
+        "llm_profile": {"type": "string", "minLength": 1}
+      }
+    },
+    "plan_annotations": {"$ref": "#/$defs/annotations"},
+    "chunks": {
+      "type": "array",
+      "minItems": 1,
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "required": ["id", "index", "source_ref", "unit_count", "annotations"],
+        "properties": {
+          "id": {"type": "string", "minLength": 1},
+          "index": {"type": "integer", "minimum": 0},
+          "source_ref": {
+            "type": "object",
+            "additionalProperties": false,
+            "required": ["source_id", "start_unit_id", "end_unit_id"],
+            "properties": {
+              "source_id": {"type": "string", "minLength": 1},
+              "start_unit_id": {"type": "integer", "minimum": 1},
+              "end_unit_id": {"type": "integer", "minimum": 1}
+            }
+          },
+          "unit_count": {"type": "integer", "minimum": 1},
+          "annotations": {"$ref": "#/$defs/annotations"}
+        }
+      }
+    }
+  },
+  "$defs": {
+    "annotations": {
+      "type": "object",
+      "propertyNames": {"type": "string", "minLength": 1},
+      "additionalProperties": true
+    }
+  }
+}

internal/framework/chunkmap/codec.go

@@ -0,0 +1,400 @@
+package chunkmap
+
+import (
+	"bytes"
+	"embed"
+	"encoding/json"
+	"fmt"
+	"io"
+	"regexp"
+	"strings"
+	"sync"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"github.com/santhosh-tekuri/jsonschema/v6"
+)
+
+//go:embed assets/schemas/source_chunk_map.v1.json
+var schemaAssets embed.FS
+
+var digestPattern = regexp.MustCompile(`^sha256:[0-9a-f]{64}$`)
+
+var (
+	loadSchemaOnce sync.Once
+	loadedSchema   []byte
+	compiledSchema *jsonschema.Schema
+	loadSchemaErr  error
+)
+
+// Codec owns strict serialization for the durable chunk-map contract.
+type Codec struct{}
+
+func New() *Codec { return &Codec{} }
+
+func (c *Codec) Kind() contracts.ArtifactKind { return ArtifactKind }
+
+func (c *Codec) Schema() contracts.ArtifactSchema {
+	raw, err := c.schemaBytes()
+	if err != nil {
+		return contracts.ArtifactSchema{}
+	}
+	return contracts.ArtifactSchema{
+		ID:         SchemaID,
+		Name:       SchemaName,
+		Version:    SchemaVersion,
+		JSONSchema: raw,
+	}
+}
+
+func (c *Codec) MediaType() string { return MediaType }
+
+// Build proves that a durable value describes the exact accepted source plan
+// and materialized chunk list supplied by the framework.
+func Build(request BuildRequest) (ChunkMap, error) {
+	if err := source.ValidateDocument(request.Source); err != nil {
+		return ChunkMap{}, fmt.Errorf("validate source document: %w", err)
+	}
+	sourceDigest, err := source.DigestDocument(request.Source)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("digest source document: %w", err)
+	}
+	if sourceDigest != request.Source.Digest {
+		return ChunkMap{}, fmt.Errorf("source digest %q does not match source document digest %q", sourceDigest, request.Source.Digest)
+	}
+	if sourceDigest != request.Plan.SourceDigest {
+		return ChunkMap{}, fmt.Errorf("source digest %q does not match chunk plan source digest %q", sourceDigest, request.Plan.SourceDigest)
+	}
+	plan, err := source.CanonicalizeChunkPlan(request.Plan)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("canonicalize chunk plan: %w", err)
+	}
+	if err := source.ValidateChunkPlan(request.Source, plan); err != nil {
+		return ChunkMap{}, fmt.Errorf("validate accepted chunk plan: %w", err)
+	}
+	planDigest, err := source.DigestChunkPlan(plan)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("digest accepted chunk plan: %w", err)
+	}
+	expected, err := source.MaterializeChunkPlan(request.Source, plan)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("materialize accepted chunk plan: %w", err)
+	}
+	if err := verifyMaterializedChunks(request.Chunks, expected); err != nil {
+		return ChunkMap{}, err
+	}
+
+	value := ChunkMap{
+		SourceID:         request.Source.ID,
+		SourceDigest:     sourceDigest,
+		PlanDigest:       planDigest,
+		RequestedChunker: request.RequestedChunker,
+		Producer:         request.Producer,
+		PlanAnnotations:  source.CloneChunkAnnotations(plan.Annotations),
+		Chunks:           make([]Chunk, len(expected)),
+	}
+	for index, chunk := range expected {
+		value.Chunks[index] = Chunk{
+			ID:          chunk.ID,
+			Index:       chunk.Index,
+			SourceRef:   chunk.Ref,
+			UnitCount:   len(chunk.Units),
+			Annotations: source.CloneChunkAnnotations(chunk.Annotations),
+		}
+	}
+	canonical, err := canonicalize(value)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("validate chunk map: %w", err)
+	}
+	return clone(canonical), nil
+}
+
+// Serialize builds and encodes the framework-owned serialized artifact.
+func Serialize(request BuildRequest) (contracts.SerializedArtifact, error) {
+	value, err := Build(request)
+	if err != nil {
+		return contracts.SerializedArtifact{}, err
+	}
+	codec := New()
+	content, err := codec.Encode(value)
+	if err != nil {
+		return contracts.SerializedArtifact{}, err
+	}
+	return contracts.SerializedArtifact{
+		Kind:      ArtifactKind,
+		Schema:    codec.Schema(),
+		MediaType: MediaType,
+		Content:   content,
+	}, nil
+}
+
+func (c *Codec) Encode(value ChunkMap) ([]byte, error) {
+	if _, err := c.schemaBytes(); err != nil {
+		return nil, err
+	}
+	canonical, err := canonicalize(clone(value))
+	if err != nil {
+		return nil, fmt.Errorf("encode source chunk map: %w", err)
+	}
+	content, err := json.Marshal(canonical)
+	if err != nil {
+		return nil, fmt.Errorf("encode source chunk map: %w", err)
+	}
+	if err := validateSchemaInstance(content); err != nil {
+		return nil, fmt.Errorf("encode source chunk map: %w", err)
+	}
+	return content, nil
+}
+
+func (c *Codec) Decode(content []byte) (ChunkMap, error) {
+	if _, err := c.schemaBytes(); err != nil {
+		return ChunkMap{}, err
+	}
+	if err := validateSchemaInstance(content); err != nil {
+		return ChunkMap{}, fmt.Errorf("decode source chunk map: %w", err)
+	}
+	decoder := json.NewDecoder(bytes.NewReader(content))
+	decoder.DisallowUnknownFields()
+	var value ChunkMap
+	if err := decoder.Decode(&value); err != nil {
+		return ChunkMap{}, fmt.Errorf("decode source chunk map: %w", err)
+	}
+	var trailing any
+	if err := decoder.Decode(&trailing); err != io.EOF {
+		return ChunkMap{}, fmt.Errorf("decode source chunk map: multiple JSON values")
+	}
+	canonical, err := canonicalize(value)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("decode source chunk map: %w", err)
+	}
+	return clone(canonical), nil
+}
+
+func (c *Codec) schemaBytes() ([]byte, error) {
+	loadSchemaOnce.Do(loadAndCompileSchema)
+	if loadSchemaErr != nil {
+		return nil, loadSchemaErr
+	}
+	return append([]byte(nil), loadedSchema...), nil
+}
+
+func loadAndCompileSchema() {
+	raw, err := schemaAssets.ReadFile("assets/schemas/source_chunk_map.v1.json")
+	if err != nil {
+		loadSchemaErr = fmt.Errorf("read source chunk map schema: %w", err)
+		return
+	}
+	var identity struct {
+		ID       string   `json:"$id"`
+		Title    string   `json:"title"`
+		Type     string   `json:"type"`
+		Required []string `json:"required"`
+	}
+	if err := json.Unmarshal(raw, &identity); err != nil {
+		loadSchemaErr = fmt.Errorf("decode source chunk map schema: %w", err)
+		return
+	}
+	if identity.ID != SchemaID || identity.Title != SchemaName || identity.Type != "object" || !hasRequiredFields(identity.Required) {
+		loadSchemaErr = fmt.Errorf("source chunk map schema identity or required fields are invalid")
+		return
+	}
+	schemaDocument, err := jsonschema.UnmarshalJSON(bytes.NewReader(raw))
+	if err != nil {
+		loadSchemaErr = fmt.Errorf("parse source chunk map schema: %w", err)
+		return
+	}
+	compiler := jsonschema.NewCompiler()
+	if err := compiler.AddResource("source-chunk-map-schema.json", schemaDocument); err != nil {
+		loadSchemaErr = fmt.Errorf("load source chunk map schema: %w", err)
+		return
+	}
+	compiled, err := compiler.Compile("source-chunk-map-schema.json")
+	if err != nil {
+		loadSchemaErr = fmt.Errorf("compile source chunk map schema: %w", err)
+		return
+	}
+	loadedSchema = append([]byte(nil), raw...)
+	compiledSchema = compiled
+}
+
+func validateSchemaInstance(content []byte) error {
+	instance, err := jsonschema.UnmarshalJSON(bytes.NewReader(content))
+	if err != nil {
+		return fmt.Errorf("payload is not valid JSON: %w", err)
+	}
+	if err := compiledSchema.Validate(instance); err != nil {
+		return fmt.Errorf("payload does not conform to source chunk map schema: %w", err)
+	}
+	return nil
+}
+
+func hasRequiredFields(required []string) bool {
+	want := map[string]bool{
+		"source_id": true, "source_digest": true, "plan_digest": true,
+		"requested_chunker": true, "producer": true, "plan_annotations": true,
+		"chunks": true,
+	}
+	for _, field := range required {
+		delete(want, field)
+	}
+	return len(want) == 0
+}
+
+func canonicalize(value ChunkMap) (ChunkMap, error) {
+	if err := requireIdentity("source_id", value.SourceID); err != nil {
+		return ChunkMap{}, err
+	}
+	if err := requireDigest("source_digest", value.SourceDigest); err != nil {
+		return ChunkMap{}, err
+	}
+	if err := requireDigest("plan_digest", value.PlanDigest); err != nil {
+		return ChunkMap{}, err
+	}
+	if err := requireIdentity("requested_chunker", value.RequestedChunker); err != nil {
+		return ChunkMap{}, err
+	}
+	if err := requireIdentity("producer.input_module", value.Producer.InputModule); err != nil {
+		return ChunkMap{}, err
+	}
+	if err := requireIdentity("producer.chunk_module", value.Producer.ChunkModule); err != nil {
+		return ChunkMap{}, err
+	}
+	if value.Producer.LLMProfile != "" {
+		if err := requireIdentity("producer.llm_profile", value.Producer.LLMProfile); err != nil {
+			return ChunkMap{}, err
+		}
+	}
+	annotations, err := canonicalizeAnnotations("plan_annotations", value.PlanAnnotations)
+	if err != nil {
+		return ChunkMap{}, err
+	}
+	value.PlanAnnotations = annotations
+	if len(value.Chunks) == 0 {
+		return ChunkMap{}, fmt.Errorf("chunks must not be empty")
+	}
+	seenIDs := make(map[string]struct{}, len(value.Chunks))
+	plan := source.ChunkPlan{SourceDigest: value.SourceDigest, Annotations: annotations, Ranges: make([]source.ChunkRange, len(value.Chunks))}
+	for index := range value.Chunks {
+		chunk := &value.Chunks[index]
+		if err := requireIdentity(fmt.Sprintf("chunks[%d].id", index), chunk.ID); err != nil {
+			return ChunkMap{}, err
+		}
+		if _, exists := seenIDs[chunk.ID]; exists {
+			return ChunkMap{}, fmt.Errorf("chunks[%d].id %q is duplicated", index, chunk.ID)
+		}
+		seenIDs[chunk.ID] = struct{}{}
+		if chunk.Index != index {
+			return ChunkMap{}, fmt.Errorf("chunks[%d].index = %d, want %d", index, chunk.Index, index)
+		}
+		if chunk.SourceRef.SourceID != value.SourceID {
+			return ChunkMap{}, fmt.Errorf("chunks[%d].source_ref.source_id %q does not match source_id %q", index, chunk.SourceRef.SourceID, value.SourceID)
+		}
+		if chunk.SourceRef.StartUnitID <= 0 || chunk.SourceRef.EndUnitID <= 0 {
+			return ChunkMap{}, fmt.Errorf("chunks[%d].source_ref endpoints must be positive", index)
+		}
+		if chunk.UnitCount <= 0 {
+			return ChunkMap{}, fmt.Errorf("chunks[%d].unit_count must be positive", index)
+		}
+		chunkAnnotations, err := canonicalizeAnnotations(fmt.Sprintf("chunks[%d].annotations", index), chunk.Annotations)
+		if err != nil {
+			return ChunkMap{}, err
+		}
+		chunk.Annotations = chunkAnnotations
+		plan.Ranges[index] = source.ChunkRange{
+			StartUnitID: chunk.SourceRef.StartUnitID,
+			EndUnitID:   chunk.SourceRef.EndUnitID,
+			Annotations: chunkAnnotations,
+		}
+	}
+	planDigest, err := source.DigestChunkPlan(plan)
+	if err != nil {
+		return ChunkMap{}, fmt.Errorf("reconstruct plan digest: %w", err)
+	}
+	if planDigest != value.PlanDigest {
+		return ChunkMap{}, fmt.Errorf("plan_digest %q does not match reconstructed plan digest %q", value.PlanDigest, planDigest)
+	}
+	return value, nil
+}
+
+func canonicalizeAnnotations(name string, annotations source.ChunkAnnotations) (source.ChunkAnnotations, error) {
+	for namespace := range annotations {
+		if strings.TrimSpace(namespace) == "" || namespace != strings.TrimSpace(namespace) {
+			return nil, fmt.Errorf("%s namespace %q must be non-empty and trimmed", name, namespace)
+		}
+	}
+	canonical, err := source.CanonicalizeChunkAnnotations(annotations)
+	if err != nil {
+		return nil, fmt.Errorf("%s: %w", name, err)
+	}
+	if canonical == nil {
+		canonical = source.ChunkAnnotations{}
+	}
+	return canonical, nil
+}
+
+func requireIdentity(name, value string) error {
+	if strings.TrimSpace(value) == "" || value != strings.TrimSpace(value) {
+		return fmt.Errorf("%s must be non-empty and trimmed", name)
+	}
+	return nil
+}
+
+func requireDigest(name, value string) error {
+	if !digestPattern.MatchString(value) {
+		return fmt.Errorf("%s must be a canonical sha256 digest", name)
+	}
+	return nil
+}
+
+func verifyMaterializedChunks(actual, expected []source.Chunk) error {
+	if len(actual) != len(expected) {
+		return fmt.Errorf("materialized chunks length = %d, want %d", len(actual), len(expected))
+	}
+	for index := range expected {
+		got, want := actual[index], expected[index]
+		if got.ID != want.ID || got.SourceID != want.SourceID || got.Index != want.Index || got.Ref != want.Ref {
+			return fmt.Errorf("materialized chunk[%d] identity or source range differs from accepted plan", index)
+		}
+		if len(got.Units) != len(want.Units) || !sameUnits(got.Units, want.Units) {
+			return fmt.Errorf("materialized chunk[%d] units differ from accepted source range", index)
+		}
+		if !sameAnnotations(got.PlanAnnotations, want.PlanAnnotations) || !sameAnnotations(got.Annotations, want.Annotations) {
+			return fmt.Errorf("materialized chunk[%d] annotations differ from accepted plan", index)
+		}
+	}
+	return nil
+}
+
+func sameUnits(left, right []source.SourceUnit) bool {
+	leftJSON, leftErr := json.Marshal(left)
+	rightJSON, rightErr := json.Marshal(right)
+	return leftErr == nil && rightErr == nil && bytes.Equal(leftJSON, rightJSON)
+}
+
+func sameAnnotations(left, right source.ChunkAnnotations) bool {
+	leftCanonical, leftErr := source.CanonicalizeChunkAnnotations(left)
+	rightCanonical, rightErr := source.CanonicalizeChunkAnnotations(right)
+	if leftErr != nil || rightErr != nil {
+		return false
+	}
+	leftJSON, leftErr := json.Marshal(leftCanonical)
+	rightJSON, rightErr := json.Marshal(rightCanonical)
+	return leftErr == nil && rightErr == nil && bytes.Equal(leftJSON, rightJSON)
+}
+
+func clone(value ChunkMap) ChunkMap {
+	value.PlanAnnotations = cloneAnnotations(value.PlanAnnotations)
+	value.Chunks = append([]Chunk(nil), value.Chunks...)
+	for index := range value.Chunks {
+		value.Chunks[index].Annotations = cloneAnnotations(value.Chunks[index].Annotations)
+	}
+	return value
+}
+
+func cloneAnnotations(annotations source.ChunkAnnotations) source.ChunkAnnotations {
+	cloned := source.CloneChunkAnnotations(annotations)
+	if cloned == nil {
+		return source.ChunkAnnotations{}
+	}
+	return cloned
+}

internal/framework/chunkmap/codec_test.go

@@ -0,0 +1,263 @@
+package chunkmap
+
+import (
+	"bytes"
+	"encoding/json"
+	"os"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+)
+
+func TestBuildAndSerializeAcceptedChunkMap(t *testing.T) {
+	request := acceptedBuildRequest(t)
+	value, err := Build(request)
+	if err != nil {
+		t.Fatalf("Build() error = %v", err)
+	}
+	if value.SourceID != request.Source.ID || len(value.Chunks) != 2 || value.Chunks[0].UnitCount != 2 || value.Chunks[1].SourceRef.StartUnitID != 20 {
+		t.Fatalf("Build() = %#v, want exact accepted chunk structure", value)
+	}
+	if value.PlanAnnotations == nil || value.Chunks[1].Annotations == nil {
+		t.Fatalf("Build() annotations = %#v, want explicit maps", value)
+	}
+	artifact, err := Serialize(request)
+	if err != nil {
+		t.Fatalf("Serialize() error = %v", err)
+	}
+	if artifact.Kind != ArtifactKind || artifact.Schema.ID != SchemaID || artifact.Schema.Name != SchemaName || artifact.Schema.Version != SchemaVersion || artifact.MediaType != MediaType || artifact.Metadata != nil {
+		t.Fatalf("Serialize() = %#v, want fixed artifact envelope without metadata", artifact)
+	}
+	decoded, err := New().Decode(artifact.Content)
+	if err != nil {
+		t.Fatalf("Decode(Serialize()) error = %v", err)
+	}
+	if decoded.PlanDigest != value.PlanDigest || decoded.Chunks[0].ID != "chunk-000001" || decoded.Chunks[1].UnitCount != 1 {
+		t.Fatalf("Decode(Serialize()) = %#v, want durable chunk map", decoded)
+	}
+}
+
+func TestCodecRoundTripsValidFixture(t *testing.T) {
+	fixture, err := os.ReadFile("testdata/source_chunk_map.v1.json")
+	if err != nil {
+		t.Fatal(err)
+	}
+	codec := New()
+	value, err := codec.Decode(fixture)
+	if err != nil {
+		t.Fatalf("Decode(fixture) error = %v", err)
+	}
+	encoded, err := codec.Encode(value)
+	if err != nil {
+		t.Fatalf("Encode(decoded fixture) error = %v", err)
+	}
+	if !bytes.Equal(encoded, bytes.TrimSpace(fixture)) {
+		t.Fatalf("fixture does not use canonical encoding\nwant: %s\n got: %s", fixture, encoded)
+	}
+}
+
+func TestBuildCanonicalizesAnnotationFormatting(t *testing.T) {
+	first := acceptedBuildRequest(t)
+	second := acceptedBuildRequest(t)
+	second.Plan.Annotations["test/chunker"] = json.RawMessage(" { \n \t\"label\" : \"fixture\" \n } ")
+	canonical, err := source.CanonicalizeChunkPlan(second.Plan)
+	if err != nil {
+		t.Fatalf("CanonicalizeChunkPlan() error = %v", err)
+	}
+	second.Chunks, err = source.MaterializeChunkPlan(second.Source, canonical)
+	if err != nil {
+		t.Fatalf("MaterializeChunkPlan() error = %v", err)
+	}
+	firstArtifact, err := Serialize(first)
+	if err != nil {
+		t.Fatalf("Serialize(first) error = %v", err)
+	}
+	secondArtifact, err := Serialize(second)
+	if err != nil {
+		t.Fatalf("Serialize(second) error = %v", err)
+	}
+	if !bytes.Equal(firstArtifact.Content, secondArtifact.Content) {
+		t.Fatalf("serialized content differs only because annotation whitespace changed\nfirst: %s\nsecond: %s", firstArtifact.Content, secondArtifact.Content)
+	}
+}
+
+func TestBuildRejectsChunksOutsideAcceptedPlan(t *testing.T) {
+	request := acceptedBuildRequest(t)
+	request.Chunks[0].Units[0].ID = 999
+	if _, err := Build(request); err == nil {
+		t.Fatal("Build() error = nil, want rejection for chunk units outside accepted source range")
+	}
+}
+
+func TestCodecRejectsInvalidDurableBoundaries(t *testing.T) {
+	value, err := Build(acceptedBuildRequest(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	for _, test := range []struct {
+		name   string
+		mutate func(*ChunkMap)
+	}{
+		{name: "blank identity", mutate: func(value *ChunkMap) { value.RequestedChunker = " " }},
+		{name: "malformed digest", mutate: func(value *ChunkMap) { value.SourceDigest = "sha256:ABC" }},
+		{name: "index mismatch", mutate: func(value *ChunkMap) { value.Chunks[1].Index = 4 }},
+		{name: "duplicate chunk id", mutate: func(value *ChunkMap) { value.Chunks[1].ID = value.Chunks[0].ID }},
+		{name: "source mismatch", mutate: func(value *ChunkMap) { value.Chunks[0].SourceRef.SourceID = "other" }},
+		{name: "invalid range", mutate: func(value *ChunkMap) { value.Chunks[0].SourceRef.StartUnitID = 0 }},
+		{name: "invalid count", mutate: func(value *ChunkMap) { value.Chunks[0].UnitCount = 0 }},
+		{name: "invalid namespace", mutate: func(value *ChunkMap) { value.PlanAnnotations[" "] = json.RawMessage(`null`) }},
+		{name: "invalid annotation", mutate: func(value *ChunkMap) { value.Chunks[0].Annotations["test/chunker"] = json.RawMessage(`{`) }},
+		{name: "plan digest mismatch", mutate: func(value *ChunkMap) { value.PlanDigest = "sha256:" + strings.Repeat("a", 64) }},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			candidate := clone(value)
+			test.mutate(&candidate)
+			if _, err := New().Encode(candidate); err == nil {
+				t.Fatal("Encode() error = nil, want invalid durable value rejection")
+			}
+		})
+	}
+	content, err := New().Encode(value)
+	if err != nil {
+		t.Fatal(err)
+	}
+	for _, raw := range [][]byte{
+		append(append([]byte(nil), content[:len(content)-1]...), []byte(`,"unknown":true}`)...),
+		append(append([]byte(nil), content...), []byte(` {}`)...),
+	} {
+		if _, err := New().Decode(raw); err == nil {
+			t.Fatalf("Decode(%s) error = nil, want strict JSON rejection", raw)
+		}
+	}
+	formatted := bytes.Replace(content, []byte(`{"label":"fixture"}`), []byte("{\n  \"label\": \"fixture\"\n}"), 1)
+	decoded, err := New().Decode(formatted)
+	if err != nil {
+		t.Fatalf("Decode(formatted annotations) error = %v", err)
+	}
+	if string(decoded.PlanAnnotations["test/chunker"]) != `{"label":"fixture"}` {
+		t.Fatalf("decoded annotation = %s, want canonical JSON", decoded.PlanAnnotations["test/chunker"])
+	}
+}
+
+func TestDecodeEnforcesRequiredSchemaFieldsAndTypes(t *testing.T) {
+	request := acceptedBuildRequest(t)
+	request.Plan.Annotations = nil
+	var err error
+	request.Chunks, err = source.MaterializeChunkPlan(request.Source, request.Plan)
+	if err != nil {
+		t.Fatal(err)
+	}
+	artifact, err := Serialize(request)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	for _, test := range []struct {
+		name   string
+		mutate func(map[string]any)
+	}{
+		{name: "missing plan annotations", mutate: func(value map[string]any) { delete(value, "plan_annotations") }},
+		{name: "null plan annotations", mutate: func(value map[string]any) { value["plan_annotations"] = nil }},
+		{name: "missing first index", mutate: func(value map[string]any) { delete(chunkDocument(value, 0), "index") }},
+		{name: "null first index", mutate: func(value map[string]any) { chunkDocument(value, 0)["index"] = nil }},
+		{name: "missing empty chunk annotations", mutate: func(value map[string]any) { delete(chunkDocument(value, 1), "annotations") }},
+		{name: "null empty chunk annotations", mutate: func(value map[string]any) { chunkDocument(value, 1)["annotations"] = nil }},
+		{name: "explicit empty llm profile", mutate: func(value map[string]any) {
+			value["producer"].(map[string]any)["llm_profile"] = ""
+		}},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			value := decodeJSONDocument(t, artifact.Content)
+			test.mutate(value)
+			content, err := json.Marshal(value)
+			if err != nil {
+				t.Fatal(err)
+			}
+			if _, err := New().Decode(content); err == nil {
+				t.Fatalf("Decode(%s) error = nil, want schema rejection", content)
+			}
+		})
+	}
+}
+
+func TestEncodeDoesNotMutateValue(t *testing.T) {
+	value, err := Build(acceptedBuildRequest(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	value.Chunks[0].Annotations["test/chunker"] = json.RawMessage(" { \n \"category\" : \"sample\" \n } ")
+	before := clone(value)
+	if _, err := New().Encode(value); err != nil {
+		t.Fatalf("Encode() error = %v", err)
+	}
+	if !reflect.DeepEqual(value, before) {
+		t.Fatalf("Encode() mutated value:\nbefore: %#v\nafter: %#v", before, value)
+	}
+}
+
+func TestChunkMapOwnershipIsIndependent(t *testing.T) {
+	request := acceptedBuildRequest(t)
+	first, err := Build(request)
+	if err != nil {
+		t.Fatal(err)
+	}
+	request.Plan.Annotations["test/chunker"][0] = '['
+	first.PlanAnnotations["test/chunker"][0] = '['
+	second, err := Build(acceptedBuildRequest(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if string(second.PlanAnnotations["test/chunker"]) != `{"label":"fixture"}` {
+		t.Fatalf("Build() shared mutable annotations: %s", second.PlanAnnotations["test/chunker"])
+	}
+}
+
+func decodeJSONDocument(t *testing.T, content []byte) map[string]any {
+	t.Helper()
+	decoder := json.NewDecoder(bytes.NewReader(content))
+	decoder.UseNumber()
+	var value map[string]any
+	if err := decoder.Decode(&value); err != nil {
+		t.Fatal(err)
+	}
+	return value
+}
+
+func chunkDocument(value map[string]any, index int) map[string]any {
+	return value["chunks"].([]any)[index].(map[string]any)
+}
+
+func acceptedBuildRequest(t *testing.T) BuildRequest {
+	t.Helper()
+	document := &source.SourceDocument{
+		ID: "source-test", Kind: "transcript", Format: "application/json",
+		Units: []source.SourceUnit{
+			{ID: 10, Kind: "segment", Text: "First unit.", Ref: source.SourceRef{SourceID: "source-test", StartUnitID: 10, EndUnitID: 10}},
+			{ID: 3, Kind: "segment", Text: "Second unit.", Ref: source.SourceRef{SourceID: "source-test", StartUnitID: 3, EndUnitID: 3}},
+			{ID: 20, Kind: "segment", Text: "Third unit.", Ref: source.SourceRef{SourceID: "source-test", StartUnitID: 20, EndUnitID: 20}},
+		},
+	}
+	digest, err := source.DigestDocument(document)
+	if err != nil {
+		t.Fatal(err)
+	}
+	document.Digest = digest
+	plan := source.ChunkPlan{
+		SourceDigest: digest,
+		Annotations:  source.ChunkAnnotations{"test/chunker": json.RawMessage(`{"label":"fixture"}`)},
+		Ranges: []source.ChunkRange{
+			{StartUnitID: 10, EndUnitID: 3, Annotations: source.ChunkAnnotations{"test/chunker": json.RawMessage(`{"category":"sample"}`)}},
+			{StartUnitID: 20, EndUnitID: 20},
+		},
+	}
+	chunks, err := source.MaterializeChunkPlan(document, plan)
+	if err != nil {
+		t.Fatal(err)
+	}
+	return BuildRequest{
+		Source: document, Plan: plan, Chunks: chunks, RequestedChunker: "chunk/requested",
+		Producer: Producer{InputModule: "input/producer", ChunkModule: "chunk/producer", LLMProfile: "profile/test"},
+	}
+}

internal/framework/chunkmap/model.go

@@ -0,0 +1,51 @@
+// Package chunkmap owns the durable accepted source chunk-map contract.
+package chunkmap
+
+import (
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+)
+
+const (
+	ArtifactKind  contracts.ArtifactKind = "source/chunk-map"
+	SchemaID                             = "notarius.source.chunk_map"
+	SchemaName                           = "notarius_source_chunk_map_v1"
+	SchemaVersion                        = "v1"
+	MediaType                            = "application/json"
+)
+
+// ChunkMap is the durable representation of one accepted materialized chunk plan.
+type ChunkMap struct {
+	SourceID         string                  `json:"source_id"`
+	SourceDigest     string                  `json:"source_digest"`
+	PlanDigest       string                  `json:"plan_digest"`
+	RequestedChunker string                  `json:"requested_chunker"`
+	Producer         Producer                `json:"producer"`
+	PlanAnnotations  source.ChunkAnnotations `json:"plan_annotations"`
+	Chunks           []Chunk                 `json:"chunks"`
+}
+
+// Producer identifies the component that produced the accepted logical plan.
+type Producer struct {
+	InputModule string `json:"input_module"`
+	ChunkModule string `json:"chunk_module"`
+	LLMProfile  string `json:"llm_profile,omitempty"`
+}
+
+// Chunk describes one accepted materialized range without source content.
+type Chunk struct {
+	ID          string                  `json:"id"`
+	Index       int                     `json:"index"`
+	SourceRef   source.SourceRef        `json:"source_ref"`
+	UnitCount   int                     `json:"unit_count"`
+	Annotations source.ChunkAnnotations `json:"annotations"`
+}
+
+// BuildRequest supplies the accepted runtime state used to build a chunk map.
+type BuildRequest struct {
+	Source           *source.SourceDocument
+	Plan             source.ChunkPlan
+	Chunks           []source.Chunk
+	RequestedChunker string
+	Producer         Producer
+}

internal/framework/chunkmap/testdata/source_chunk_map.v1.json

@@ -0,0 +1 @@
+{"source_id":"source-test","source_digest":"sha256:186b2d30029e7fda40545f88e75ade38f22bff3e5543af4dfeb86587536a01af","plan_digest":"sha256:e50cc7da9070ac8a4339d79b2c206be842846c5c22af7b0ad58f7e9a34ceaf97","requested_chunker":"chunk/requested","producer":{"input_module":"input/producer","chunk_module":"chunk/producer","llm_profile":"profile/test"},"plan_annotations":{"test/chunker":{"label":"fixture"}},"chunks":[{"id":"chunk-000001","index":0,"source_ref":{"source_id":"source-test","start_unit_id":10,"end_unit_id":3},"unit_count":2,"annotations":{"test/chunker":{"category":"sample"}}},{"id":"chunk-000002","index":1,"source_ref":{"source_id":"source-test","start_unit_id":20,"end_unit_id":20},"unit_count":1,"annotations":{}}]}

internal/framework/chunkplan/store_test.go

@@ -281,7 +281,7 @@ func TestFilesystemStoreReportsInvalidRecordsAsRecoverable(t *testing.T) {
 			return bytes.Replace(data, []byte(`{"schema_version"`), []byte(`{"SENTINEL_UNKNOWN_FIELD":true,"schema_version"`), 1)
 		}},
 		{name: "truncated JSON", mutate: func(data []byte) []byte { return data[:len(data)/2] }},
-		{name: "schema mismatch", mutate: replaceJSON(`notarius.chunk-plan.v1`, `SENTINEL_SCHEMA_VALUE`)},
+		{name: "legacy v1 record", mutate: replaceJSON(`notarius.chunk-plan.v2`, `notarius.chunk-plan.v1`)},
 		{name: "source mismatch", mutate: replaceJSON(testSourceDigest, "sha256:"+strings.Repeat("b", 64))},
 		{name: "plan digest mismatch", mutate: func(data []byte) []byte {
 			prefix := []byte(`"plan_digest":"sha256:`)

internal/framework/contracts/artifact.go

@@ -69,6 +69,16 @@ func CloneSerializedArtifact(artifact SerializedArtifact) SerializedArtifact {
 	return artifact
 }
 
+// CloneSerializedArtifactPointer returns an independently owned artifact when
+// one is present.
+func CloneSerializedArtifactPointer(artifact *SerializedArtifact) *SerializedArtifact {
+	if artifact == nil {
+		return nil
+	}
+	cloned := CloneSerializedArtifact(*artifact)
+	return &cloned
+}
+
 func CloneSerializedOutput(output SerializedOutput) SerializedOutput {
 	output.Artifact = CloneSerializedArtifact(output.Artifact)
 	return output

internal/framework/contracts/contracts.go

@@ -291,6 +291,7 @@ type OutputRequest struct {
 	Warnings         []Warning             `json:"warnings,omitempty"`
 	LLMProfile       string                `json:"llm_profile,omitempty"`
 	Metadata         map[string]any        `json:"metadata,omitempty"`
+	ChunkMap         *SerializedArtifact   `json:"chunk_map,omitempty"`
 }
 
 type OutputFile struct {

internal/framework/contracts/contracts_test.go

@@ -26,6 +26,25 @@ func TestCloneReferenceSlotsEmptyInputReturnsNil(t *testing.T) {
 	}
 }
 
+func TestCloneSerializedArtifactPointerOwnsArtifactData(t *testing.T) {
+	original := &SerializedArtifact{
+		Kind:     "test/artifact",
+		Schema:   ArtifactSchema{ID: "test.schema", JSONSchema: []byte(`{"type":"object"}`)},
+		Content:  []byte(`{"items":[]}`),
+		Metadata: map[string]any{"count": 1},
+	}
+	cloned := CloneSerializedArtifactPointer(original)
+	original.Schema.JSONSchema[0] = '['
+	original.Content[0] = '['
+	original.Metadata["count"] = 2
+	if cloned == nil || string(cloned.Schema.JSONSchema) != `{"type":"object"}` || string(cloned.Content) != `{"items":[]}` || cloned.Metadata["count"] != 1 {
+		t.Fatalf("CloneSerializedArtifactPointer() = %#v, want independent artifact data", cloned)
+	}
+	if CloneSerializedArtifactPointer(nil) != nil {
+		t.Fatal("CloneSerializedArtifactPointer(nil) must return nil")
+	}
+}
+
 func TestCloneReferenceSlotsPreservesFields(t *testing.T) {
 	slots := []ReferenceSlot{
 		{

internal/framework/pipeline/chunk_plan_store.go

@@ -8,7 +8,7 @@ import (
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
 )
 
-const ChunkPlanSchemaVersion = "notarius.chunk-plan.v1"
+const ChunkPlanSchemaVersion = "notarius.chunk-plan.v2"
 
 type ChunkPlanProducer struct {
 	InputModule string                          `json:"input_module"`

internal/framework/pipeline/runner.go

@@ -16,6 +16,7 @@ import (
 
 	"gitea.maximumdirect.net/eric/notarius/internal/core/artifacts"
 	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/chunkmap"
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
 )
 
@@ -251,6 +252,28 @@ func (r *Runner) Run(ctx context.Context, input RunInput) (output RunOutput, err
 		return failOutput(output), fmt.Errorf("write chunk debug artifact: %w", err)
 	}
 
+	var acceptedChunkMap *contracts.SerializedArtifact
+	if chunkResult.accepted {
+		if chunkResult.plan == nil || chunkResult.record == nil {
+			return failOutput(output), fmt.Errorf("accepted chunk plan is missing plan or producer record")
+		}
+		artifact, buildErr := chunkmap.Serialize(chunkmap.BuildRequest{
+			Source:           doc,
+			Plan:             *chunkResult.plan,
+			Chunks:           chunkResult.chunks,
+			RequestedChunker: input.pipeline.Chunk.Module,
+			Producer: chunkmap.Producer{
+				InputModule: chunkResult.record.Producer.InputModule,
+				ChunkModule: chunkResult.record.Producer.ChunkModule,
+				LLMProfile:  chunkResult.record.Producer.LLMProfile,
+			},
+		})
+		if buildErr != nil {
+			return failOutput(output), fmt.Errorf("serialize accepted chunk map: %w", buildErr)
+		}
+		acceptedChunkMap = &artifact
+	}
+
 	if chunkResult.accepted {
 		if err := r.runPreparedSteps(ctx, input, checkpoints, checkpointLoader, doc, sourceInput, sessionID, chunkResult.chunks, &output); err != nil {
 			return failOutput(output), err
@@ -270,18 +293,22 @@ func (r *Runner) Run(ctx context.Context, input RunInput) (output RunOutput, err
 		return failOutput(output), err
 	}
 	outputStarted := time.Now().UTC()
-	if err := writeDebugTimed(debugRecorder, "output/input.json", debugTimedEnvelope{
-		Stage:     string(StageOutput),
-		ModuleKey: encoder.Key(),
-		StartedAt: outputStarted,
-		Payload: map[string]any{
+	outputDebugPayload := map[string]any{
 		"manifest":          output.Manifest,
 		"normalize_outputs": debugSerializedOutputEnvelopes(output.NormalizeOutputs),
 		"rejected":          debugRejectedOutputEnvelopes(output.Rejected),
 		"warnings":          output.Warnings,
 		"options":           redactSensitiveMap(input.pipeline.Output.Options),
 		"metadata":          redactSensitiveMap(input.Metadata),
-		},
+	}
+	if acceptedChunkMap != nil {
+		outputDebugPayload["chunk_map"] = debugSerializedOutputEnvelope(contracts.SerializedOutput{Artifact: *acceptedChunkMap})
+	}
+	if err := writeDebugTimed(debugRecorder, "output/input.json", debugTimedEnvelope{
+		Stage:     string(StageOutput),
+		ModuleKey: encoder.Key(),
+		StartedAt: outputStarted,
+		Payload:   outputDebugPayload,
 	}); err != nil {
 		return failOutput(output), fmt.Errorf("write output debug artifact: %w", err)
 	}
@@ -296,6 +323,7 @@ func (r *Runner) Run(ctx context.Context, input RunInput) (output RunOutput, err
 		Warnings:         output.Warnings,
 		LLMProfile:       input.pipeline.Output.LLMProfile,
 		Metadata:         outputMetadata,
+		ChunkMap:         contracts.CloneSerializedArtifactPointer(acceptedChunkMap),
 	})
 	output.Warnings = append(output.Warnings, encoded.Warnings...)
 	if err != nil {

internal/framework/pipeline/runner_chunk_plan_test.go

@@ -10,6 +10,7 @@ import (
 	"time"
 
 	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/chunkmap"
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
 )
 
@@ -40,6 +41,17 @@ type countingChunkValidator struct {
 	err    error
 }
 
+type capturingChunkMapOutput struct {
+	requests []contracts.OutputRequest
+}
+
+func (*capturingChunkMapOutput) Key() string { return "capture/chunk-map" }
+
+func (output *capturingChunkMapOutput) Encode(_ context.Context, request contracts.OutputRequest) (contracts.OutputResult, error) {
+	output.requests = append(output.requests, request)
+	return contracts.OutputResult{}, nil
+}
+
 func (*countingChunkValidator) Name() string { return "test/counting-chunks" }
 func (*countingChunkValidator) ExecutionClass() contracts.ExecutionClass {
 	return contracts.ExecutionClassDeterministic
@@ -205,6 +217,116 @@ func TestRunnerChunkPlanHitUsesStoredProducerProvenance(t *testing.T) {
 	}
 }
 
+func TestRunnerProvidesAcceptedChunkMapToOutput(t *testing.T) {
+	for _, test := range []struct {
+		name      string
+		configure func(*PreparedPipeline, source.ChunkPlan) (RunInput, chunkmap.Producer)
+	}{
+		{
+			name: "bypassed plan",
+			configure: func(prepared *PreparedPipeline, _ source.ChunkPlan) (RunInput, chunkmap.Producer) {
+				return RunInput{Prepared: prepared, RawInput: []byte("input"), ChunkCacheMode: ChunkCacheBypass}, chunkmap.Producer{
+					InputModule: prepared.input.Key(), ChunkModule: prepared.chunker.Key(),
+				}
+			},
+		},
+		{
+			name: "cached plan",
+			configure: func(prepared *PreparedPipeline, plan source.ChunkPlan) (RunInput, chunkmap.Producer) {
+				prepared.resolved.Chunk.Module = "chunk/requested"
+				record := chunkPlanRecord(t, prepared, plan)
+				record.Producer.InputModule = "input/original"
+				record.Producer.ChunkModule = "chunk/original"
+				record.Producer.LLMProfile = "original-profile"
+				return RunInput{
+					Prepared: prepared, RawInput: []byte("input"), ChunkCacheMode: ChunkCacheAuto,
+					ChunkPlans: &recordingChunkPlanStore{record: record, decision: ChunkPlanDecision{Status: ChunkPlanHit}},
+				}, chunkmap.Producer{InputModule: "input/original", ChunkModule: "chunk/original", LLMProfile: "original-profile"}
+			},
+		},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			prepared, plan := preparedTerminalDebugPipeline(t)
+			encoder := &capturingChunkMapOutput{}
+			prepared.output = encoder
+			input, wantProducer := test.configure(prepared, plan)
+			if _, err := New().Run(context.Background(), input); err != nil {
+				t.Fatalf("Run() error = %v", err)
+			}
+			if len(encoder.requests) != 1 || encoder.requests[0].ChunkMap == nil {
+				t.Fatalf("output requests = %#v, want one accepted chunk map", encoder.requests)
+			}
+			artifact := encoder.requests[0].ChunkMap
+			if artifact.Kind != chunkmap.ArtifactKind || artifact.Schema.ID != chunkmap.SchemaID || artifact.MediaType != chunkmap.MediaType {
+				t.Fatalf("chunk map artifact = %#v, want fixed serialized identity", artifact)
+			}
+			value, err := chunkmap.New().Decode(artifact.Content)
+			if err != nil {
+				t.Fatalf("decode output chunk map: %v", err)
+			}
+			if value.RequestedChunker != prepared.resolved.Chunk.Module || value.Producer != wantProducer {
+				t.Fatalf("chunk map = %#v, want current request and producer %#v", value, wantProducer)
+			}
+		})
+	}
+}
+
+func TestRunnerOmitsChunkMapForRejectedPlan(t *testing.T) {
+	prepared, _ := preparedTerminalDebugPipeline(t)
+	encoder := &capturingChunkMapOutput{}
+	prepared.output = encoder
+	prepared.chunkValidators.validators = []preparedValidator{{
+		resolved: ResolvedValidator{Binding: Binding("reject/chunk"), Target: ValidatorTargetChunk},
+		chunk:    &countingChunkValidator{result: contracts.ValidationResult{Approved: false, ReasonCode: "rejected", Message: "not accepted"}},
+	}}
+	output, err := New().Run(context.Background(), RunInput{Prepared: prepared, RawInput: []byte("input")})
+	if err != nil {
+		t.Fatalf("Run() error = %v", err)
+	}
+	if len(output.Rejected) != 1 || len(encoder.requests) != 1 || encoder.requests[0].ChunkMap != nil {
+		t.Fatalf("output = %#v requests = %#v, want rejected plan without chunk map", output.Rejected, encoder.requests)
+	}
+}
+
+func TestRunnerRetainsChunkMapAfterLaneRejection(t *testing.T) {
+	prepared, _ := preparedTerminalDebugPipeline(t)
+	encoder := &capturingChunkMapOutput{}
+	prepared.output = encoder
+	prepared.Steps[0].lanes[0].mergeValidators.validators = []preparedValidator{{
+		resolved: ResolvedValidator{Binding: Binding("reject/lane"), Target: ValidatorTargetTyped, ArtifactKind: "test/notes"},
+		typedValidate: func(context.Context, any, typedValidationTarget) (contracts.ValidationResult, error) {
+			return contracts.ValidationResult{Approved: false, ReasonCode: "rejected", Message: "not accepted"}, nil
+		},
+	}}
+	output, err := New().Run(context.Background(), RunInput{Prepared: prepared, RawInput: []byte("input")})
+	if err != nil {
+		t.Fatalf("Run() error = %v", err)
+	}
+	if len(output.Rejected) != 1 || len(encoder.requests) != 1 || encoder.requests[0].ChunkMap == nil {
+		t.Fatalf("output = %#v requests = %#v, want lane rejection with accepted chunk map", output.Rejected, encoder.requests)
+	}
+}
+
+func TestRunnerChunkMapRequestDoesNotAliasStoredPlan(t *testing.T) {
+	prepared, plan := preparedTerminalDebugPipeline(t)
+	encoder := &capturingChunkMapOutput{}
+	prepared.output = encoder
+	record := chunkPlanRecord(t, prepared, plan)
+	record.Plan.Annotations = source.ChunkAnnotations{"dnd/scenes": json.RawMessage(`{"title":"Opening"}`)}
+	store := &recordingChunkPlanStore{record: record, decision: ChunkPlanDecision{Status: ChunkPlanHit}}
+	if _, err := New().Run(context.Background(), RunInput{Prepared: prepared, RawInput: []byte("input"), ChunkCacheMode: ChunkCacheAuto, ChunkPlans: store}); err != nil {
+		t.Fatalf("Run() error = %v", err)
+	}
+	store.record.Plan.Annotations["dnd/scenes"][0] = '['
+	value, err := chunkmap.New().Decode(encoder.requests[0].ChunkMap.Content)
+	if err != nil {
+		t.Fatalf("decode captured chunk map after store mutation: %v", err)
+	}
+	if string(value.PlanAnnotations["dnd/scenes"]) != `{"title":"Opening"}` {
+		t.Fatalf("captured chunk map aliases stored plan: %s", value.PlanAnnotations["dnd/scenes"])
+	}
+}
+
 func TestRunnerChunkPlanManifestRetainsCandidateOnRejection(t *testing.T) {
 	prepared, _ := preparedTerminalDebugPipeline(t)
 	validator := &countingChunkValidator{result: contracts.ValidationResult{Approved: false, ReasonCode: "policy", Message: "no"}}
@@ -280,19 +402,19 @@ func TestRunnerRegeneratesStructurallyInvalidHit(t *testing.T) {
 	}
 }
 
-func TestRunnerReusesCrossDomainAnnotationsAsOptionalData(t *testing.T) {
+func TestRunnerReusesAnnotationsFromDifferentChunkerAsOptionalData(t *testing.T) {
 	prepared, plan := preparedTerminalDebugPipeline(t)
-	plan.Annotations = source.ChunkAnnotations{"dnd/scenes": json.RawMessage(`{"boundary_caveats":["uncertain"]}`)}
-	plan.Ranges[0].Annotations = source.ChunkAnnotations{"dnd/scenes": json.RawMessage(`{"title":"Opening"}`)}
+	plan.Annotations = source.ChunkAnnotations{"test/chunker": json.RawMessage(`{"label":"fixture"}`)}
+	plan.Ranges[0].Annotations = source.ChunkAnnotations{"test/chunker": json.RawMessage(`{"category":"sample"}`)}
 	record := chunkPlanRecord(t, prepared, plan)
-	record.Producer.ChunkModule = "dnd/scenes"
+	record.Producer.ChunkModule = "chunk/producer"
 	store := &recordingChunkPlanStore{record: record, decision: ChunkPlanDecision{Status: ChunkPlanHit}}
 	output, err := New().Run(context.Background(), RunInput{Prepared: prepared, RawInput: []byte("input"), ChunkCacheMode: ChunkCacheAuto, ChunkPlans: store})
 	if err != nil {
 		t.Fatal(err)
 	}
-	if output.Manifest.ChunkPlan.ProducerModule != "dnd/scenes" || output.Manifest.ChunkPlan.Action != "reused" {
-		t.Fatalf("cross-domain annotation plan was not reused: %#v", output.Manifest.ChunkPlan)
+	if output.Manifest.ChunkPlan.ProducerModule != "chunk/producer" || output.Manifest.ChunkPlan.Action != "reused" {
+		t.Fatalf("different-chunker annotation plan was not reused: %#v", output.Manifest.ChunkPlan)
 	}
 }
 

internal/modules/dnd/chunk/scenes/assets/prompts/instructions.md

@@ -1,52 +1,8 @@
-Good reasons to start a new scene include:
-- the party moves to a new location;
-- a combat encounter begins or ends;
-- combat changes into a substantially different phase;
-- the party shifts between combat, exploration, social interaction, discussion,
-  planning, travel, rest, or downtime;
-- a new NPC, faction, threat, or objective becomes central;
-- the party completes one immediate goal and begins another;
-- a major table-level rules discussion interrupts and materially changes play.
+Cover the complete provided transcript from its first source unit to its last
+source unit. Return scenes in source-unit order with no gaps or overlaps. Use
+only positive integer source-unit IDs from the transcript, and give every scene
+one inclusive `start_unit_id` and one inclusive `end_unit_id`.
 
-Do not start a new scene merely because:
-- the speaker changes;
-- a new combat round begins;
-- a player asks a brief rules question;
-- there is a joke, aside, or short table comment;
-- a character takes a routine turn;
-- the same encounter continues without a meaningful change in situation.
-
-dnd/scenes boundary policy:
-- cover the full provided transcript from the first source unit to the last
-  source unit;
-- return sequential scenes with no gaps;
-- do not overlap scenes;
-- preserve source-unit order;
-- use integer source-unit IDs from the transcript;
-- each scene must have start_unit_id and end_unit_id;
-- do not include final chunk IDs or chunk indexes.
-
-For each scene:
-- short_title should be brief and factual;
-- primary_mode must be Recap, Discussion, Combat, or Narrative;
-- main_participants should include only principal characters, NPCs, factions, or
-  groups involved;
-- summary should be factual and compact, usually one to three sentences;
-- boundary_note should explain why the scene begins at start_unit_id and ends at
-  end_unit_id;
-- boundary_confidence must be High, Medium, or Low.
-
-Primary mode guidance:
-- Use Recap for opening recap, initiative setup, session framing, or immediate
-  continuation from prior events.
-- Use Discussion when the party is primarily discussing options or choosing a
-  course of action.
-- Use Combat when active combat or combat-resolution mechanics dominate.
-- Use Narrative for all other non-combat gameplay, including exploration, social
-  interactions, shopping, preparation, travel, rest, and downtime.
-
-In boundary_caveats, list overall caveats about scene divisions. Include scenes
-that could reasonably be split differently, combat phases that were kept
-together, gradual transitions, or places where map context would have helped.
-
-Return exactly one JSON object and no explanatory text.
+Return exactly one JSON object and no explanatory text. The object must contain
+only a non-empty `scenes` array. Each scene object must contain only
+`start_unit_id` and `end_unit_id`.

internal/modules/dnd/chunk/scenes/assets/prompts/task.md

@@ -1,5 +1,18 @@
 Divide the provided transcript into coherent Dungeons & Dragons scenes for the
-dnd/scenes chunk module.
+`dnd/scenes` chunk module.
 
-A scene is a coherent unit of play. Start a new scene when there is a meaningful
-change in location, objective, threat, activity, encounter, or mode of play.
+A scene is a coherent unit of play. Start a new scene when the transcript
+establishes a meaningful change in location, objective, threat, activity,
+encounter, or mode of play. Good reasons include a material move, beginning or
+ending combat, a substantially different encounter phase, a shift between
+combat, exploration, social interaction, planning, travel, rest, or downtime,
+a change in the central NPC, faction, threat, or objective, or a sustained
+table-level interruption that materially changes the activity.
+
+Do not split a scene merely because a speaker or combat round changes, a
+routine turn occurs, or the table briefly digresses. Prefer fewer coherent
+scenes over speculative or fine-grained boundaries.
+
+Return only inclusive `start_unit_id` and `end_unit_id` endpoints for each
+scene. Do not return titles, modes, participants, summaries, boundary notes,
+confidence, caveats, final chunk IDs, or chunk indexes.

internal/modules/dnd/chunk/scenes/assets/schemas/dnd_scenes.v1.json

@@ -3,10 +3,7 @@
   "$id": "notarius.dnd.scenes",
   "type": "object",
   "additionalProperties": false,
-  "required": [
-    "scenes",
-    "boundary_caveats"
-  ],
+  "required": ["scenes"],
   "properties": {
     "scenes": {
       "type": "array",
@@ -14,16 +11,7 @@
       "items": {
         "type": "object",
         "additionalProperties": false,
-        "required": [
-          "start_unit_id",
-          "end_unit_id",
-          "short_title",
-          "primary_mode",
-          "main_participants",
-          "summary",
-          "boundary_note",
-          "boundary_confidence"
-        ],
+        "required": ["start_unit_id", "end_unit_id"],
         "properties": {
           "start_unit_id": {
             "type": "integer",
@@ -32,53 +20,9 @@
           "end_unit_id": {
             "type": "integer",
             "minimum": 1
-          },
-          "short_title": {
-            "type": "string",
-            "minLength": 1
-          },
-          "primary_mode": {
-            "type": "string",
-            "enum": [
-              "Recap",
-              "Discussion",
-              "Combat",
-              "Narrative"
-            ]
-          },
-          "main_participants": {
-            "type": "array",
-            "minItems": 1,
-            "items": {
-              "type": "string",
-              "minLength": 1
-            }
-          },
-          "summary": {
-            "type": "string",
-            "minLength": 1
-          },
-          "boundary_note": {
-            "type": "string",
-            "minLength": 1
-          },
-          "boundary_confidence": {
-            "type": "string",
-            "enum": [
-              "High",
-              "Medium",
-              "Low"
-            ]
           }
         }
       }
-    },
-    "boundary_caveats": {
-      "type": "array",
-      "items": {
-        "type": "string",
-        "minLength": 1
-      }
     }
   }
 }

internal/modules/dnd/chunk/scenes/chunker.go

@@ -2,9 +2,7 @@ package scenes
 
 import (
 	"context"
-	"encoding/json"
 	"fmt"
-	"strings"
 
 	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
@@ -22,8 +20,6 @@ var providedCapabilities = []string{
 	"chunks",
 }
 
-const annotationNamespace = "dnd/scenes"
-
 var referenceSlotDescriptions = shared.ReferenceSlotDescriptions{
 	Glossary: "Optional campaign glossary reference material used only for scene disambiguation.",
 	Party:    "Optional party roster reference material used only for scene disambiguation.",
@@ -113,18 +109,11 @@ func (c *Chunker) Plan(ctx context.Context, req contracts.ChunkRequest) (contrac
 		return contracts.ChunkPlanResult{}, chunkerErrorf("complete structured output: %w", err)
 	}
 
-	warnings, err := warningsFromCaveats(response.BoundaryCaveats)
+	plan, err := planFromResponse(req.Source, response)
 	if err != nil {
 		return contracts.ChunkPlanResult{}, chunkerErrorf("malformed structured output: %w", err)
 	}
-	plan, err := planFromResponse(req.Source, response, warnings)
-	if err != nil {
-		return contracts.ChunkPlanResult{}, chunkerErrorf("malformed structured output: %w", err)
-	}
-	return contracts.ChunkPlanResult{
-		Plan:     plan,
-		Warnings: warnings,
-	}, nil
+	return contracts.ChunkPlanResult{Plan: plan}, nil
 }
 
 func ModuleSpec() pipeline.ModuleSpec {
@@ -159,7 +148,7 @@ func DecodeOptions(options map[string]any) (Options, error) {
 	return Options{}, nil
 }
 
-func planFromResponse(doc *source.SourceDocument, response chunkResponse, warnings []contracts.Warning) (source.ChunkPlan, error) {
+func planFromResponse(doc *source.SourceDocument, response chunkResponse) (source.ChunkPlan, error) {
 	if response.Scenes == nil {
 		return source.ChunkPlan{}, fmt.Errorf("scenes must be present")
 	}
@@ -175,21 +164,25 @@ func planFromResponse(doc *source.SourceDocument, response chunkResponse, warnin
 	ranges := make([]source.ChunkRange, 0, len(response.Scenes))
 	previousEnd := -1
 	for i, scene := range response.Scenes {
-		normalized, err := normalizeScene(doc, i, scene)
+		startUnitID, err := shared.ResolveUnitID(doc, "start_unit_id", scene.StartUnitID)
 		if err != nil {
-			return source.ChunkPlan{}, err
+			return source.ChunkPlan{}, fmt.Errorf("scene[%d] %w", i, err)
+		}
+		endUnitID, err := shared.ResolveUnitID(doc, "end_unit_id", scene.EndUnitID)
+		if err != nil {
+			return source.ChunkPlan{}, fmt.Errorf("scene[%d] %w", i, err)
 		}
 
-		startIndex, ok := unitIndexes[normalized.StartUnitID]
+		startIndex, ok := unitIndexes[startUnitID]
 		if !ok {
-			return source.ChunkPlan{}, fmt.Errorf("scene[%d] start_unit_id %d was not found", i, normalized.StartUnitID)
+			return source.ChunkPlan{}, fmt.Errorf("scene[%d] start_unit_id %d was not found", i, startUnitID)
 		}
-		endIndex, ok := unitIndexes[normalized.EndUnitID]
+		endIndex, ok := unitIndexes[endUnitID]
 		if !ok {
-			return source.ChunkPlan{}, fmt.Errorf("scene[%d] end_unit_id %d was not found", i, normalized.EndUnitID)
+			return source.ChunkPlan{}, fmt.Errorf("scene[%d] end_unit_id %d was not found", i, endUnitID)
 		}
 		if startIndex > endIndex {
-			return source.ChunkPlan{}, fmt.Errorf("scene[%d] start_unit_id %d appears after end_unit_id %d", i, normalized.StartUnitID, normalized.EndUnitID)
+			return source.ChunkPlan{}, fmt.Errorf("scene[%d] start_unit_id %d appears after end_unit_id %d", i, startUnitID, endUnitID)
 		}
 
 		if i == 0 && startIndex != 0 {
@@ -205,142 +198,21 @@ func planFromResponse(doc *source.SourceDocument, response chunkResponse, warnin
 		}
 		previousEnd = endIndex
 
-		annotation, err := json.Marshal(struct {
-			ShortTitle         string   `json:"short_title"`
-			PrimaryMode        string   `json:"primary_mode"`
-			MainParticipants   []string `json:"main_participants"`
-			Summary            string   `json:"summary"`
-			BoundaryNote       string   `json:"boundary_note"`
-			BoundaryConfidence string   `json:"boundary_confidence"`
-		}{normalized.ShortTitle, normalized.PrimaryMode, normalized.MainParticipants, normalized.Summary, normalized.BoundaryNote, normalized.BoundaryConfidence})
-		if err != nil {
-			return source.ChunkPlan{}, fmt.Errorf("encode scene[%d] annotation: %w", i, err)
-		}
 		ranges = append(ranges, source.ChunkRange{
-			StartUnitID: normalized.StartUnitID,
-			EndUnitID:   normalized.EndUnitID,
-			Annotations: source.ChunkAnnotations{annotationNamespace: annotation},
+			StartUnitID: startUnitID,
+			EndUnitID:   endUnitID,
 		})
 	}
 
 	if previousEnd != len(doc.Units)-1 {
 		return source.ChunkPlan{}, fmt.Errorf("final scene must end at final source unit %d", doc.Units[len(doc.Units)-1].ID)
 	}
-	caveats := make([]string, 0, len(warnings))
-	for _, warning := range warnings {
-		caveats = append(caveats, warning.Message)
-	}
-	annotation, err := json.Marshal(struct {
-		BoundaryCaveats []string `json:"boundary_caveats"`
-	}{BoundaryCaveats: caveats})
-	if err != nil {
-		return source.ChunkPlan{}, fmt.Errorf("encode plan annotation: %w", err)
-	}
 	return source.ChunkPlan{
 		SourceDigest: doc.Digest,
 		Ranges:       ranges,
-		Annotations:  source.ChunkAnnotations{annotationNamespace: annotation},
 	}, nil
 }
 
-func normalizeScene(doc *source.SourceDocument, index int, scene sceneResponse) (normalizedScene, error) {
-	startUnitID, err := shared.ResolveUnitID(doc, "start_unit_id", scene.StartUnitID)
-	if err != nil {
-		return normalizedScene{}, fmt.Errorf("scene[%d] %w", index, err)
-	}
-	endUnitID, err := shared.ResolveUnitID(doc, "end_unit_id", scene.EndUnitID)
-	if err != nil {
-		return normalizedScene{}, fmt.Errorf("scene[%d] %w", index, err)
-	}
-
-	out := normalizedScene{
-		StartUnitID:        startUnitID,
-		EndUnitID:          endUnitID,
-		ShortTitle:         strings.TrimSpace(scene.ShortTitle),
-		PrimaryMode:        strings.TrimSpace(scene.PrimaryMode),
-		Summary:            strings.TrimSpace(scene.Summary),
-		BoundaryNote:       strings.TrimSpace(scene.BoundaryNote),
-		BoundaryConfidence: strings.TrimSpace(scene.BoundaryConfidence),
-	}
-
-	requiredInts := map[string]int{
-		"start_unit_id": out.StartUnitID,
-		"end_unit_id":   out.EndUnitID,
-	}
-	for field, value := range requiredInts {
-		if value <= 0 {
-			return normalizedScene{}, fmt.Errorf("scene[%d] %s must be positive", index, field)
-		}
-	}
-	required := map[string]string{
-		"short_title":         out.ShortTitle,
-		"primary_mode":        out.PrimaryMode,
-		"summary":             out.Summary,
-		"boundary_note":       out.BoundaryNote,
-		"boundary_confidence": out.BoundaryConfidence,
-	}
-	for field, value := range required {
-		if value == "" {
-			return normalizedScene{}, fmt.Errorf("scene[%d] %s must not be empty", index, field)
-		}
-	}
-	if !validPrimaryMode(out.PrimaryMode) {
-		return normalizedScene{}, fmt.Errorf("scene[%d] primary_mode %q is not supported", index, out.PrimaryMode)
-	}
-	if !validBoundaryConfidence(out.BoundaryConfidence) {
-		return normalizedScene{}, fmt.Errorf("scene[%d] boundary_confidence %q is not supported", index, out.BoundaryConfidence)
-	}
-	if len(scene.MainParticipants) == 0 {
-		return normalizedScene{}, fmt.Errorf("scene[%d] main_participants must not be empty", index)
-	}
-	out.MainParticipants = make([]string, 0, len(scene.MainParticipants))
-	for participantIndex, participant := range scene.MainParticipants {
-		trimmed := strings.TrimSpace(participant)
-		if trimmed == "" {
-			return normalizedScene{}, fmt.Errorf("scene[%d] main_participants[%d] must not be empty", index, participantIndex)
-		}
-		out.MainParticipants = append(out.MainParticipants, trimmed)
-	}
-	return out, nil
-}
-
-func validPrimaryMode(value string) bool {
-	switch value {
-	case "Recap", "Discussion", "Combat", "Narrative":
-		return true
-	default:
-		return false
-	}
-}
-
-func validBoundaryConfidence(value string) bool {
-	switch value {
-	case "High", "Medium", "Low":
-		return true
-	default:
-		return false
-	}
-}
-
-func warningsFromCaveats(caveats []string) ([]contracts.Warning, error) {
-	if len(caveats) == 0 {
-		return nil, nil
-	}
-	warnings := make([]contracts.Warning, 0, len(caveats))
-	for i, caveat := range caveats {
-		trimmed := strings.TrimSpace(caveat)
-		if trimmed == "" {
-			return nil, fmt.Errorf("boundary_caveats[%d] must not be empty after trimming", i)
-		}
-		warnings = append(warnings, contracts.Warning{
-			Scope:      Key,
-			ReasonCode: "scene_boundary_caveat",
-			Message:    trimmed,
-		})
-	}
-	return warnings, nil
-}
-
 func chunkerErrorf(format string, args ...any) error {
 	return fmt.Errorf("dnd scenes chunker: "+format, args...)
 }

internal/modules/dnd/chunk/scenes/chunker_test.go

@@ -110,32 +110,13 @@ func wantReferenceSlots() []contracts.ReferenceSlot {
 	}
 }
 
-func TestPlanReturnsSceneRangesAndAnnotationsFromStructuredOutput(t *testing.T) {
+func TestPlanReturnsAnnotationFreeSceneRangesFromStructuredOutput(t *testing.T) {
 	client := &fakeScenesLLMClient{
 		response: chunkResponse{
 			Scenes: []sceneResponse{
-				{
-					StartUnitID:        shared.UnitRefFromInt(1),
-					EndUnitID:          shared.UnitRefFromInt(2),
-					ShortTitle:         " Goblin parley ",
-					PrimaryMode:        "Discussion",
-					MainParticipants:   []string{" Aria ", "Goblin scout"},
-					Summary:            " The party negotiates with a scout. ",
-					BoundaryNote:       " The scene covers the discussion before fighting starts. ",
-					BoundaryConfidence: "High",
+				scene(1, 2),
+				scene(3, 4),
 			},
-				{
-					StartUnitID:        shared.UnitRefFromInt(3),
-					EndUnitID:          shared.UnitRefFromInt(4),
-					ShortTitle:         "Ambush at the gate",
-					PrimaryMode:        "Combat",
-					MainParticipants:   []string{"Aria", "Goblin ambushers"},
-					Summary:            "The goblins attack at the gate.",
-					BoundaryNote:       "Combat begins and resolves the immediate threat.",
-					BoundaryConfidence: "Medium",
-				},
-			},
-			BoundaryCaveats: []string{" The transition into combat is gradual. "},
 		},
 	}
 
@@ -185,55 +166,67 @@ func TestPlanReturnsSceneRangesAndAnnotationsFromStructuredOutput(t *testing.T)
 		t.Fatalf("ranges = %#v, want two", result.Plan.Ranges)
 	}
 	for i, want := range wantRanges {
-		if result.Plan.Ranges[i].StartUnitID != want.StartUnitID || result.Plan.Ranges[i].EndUnitID != want.EndUnitID {
+		got := result.Plan.Ranges[i]
+		if got.StartUnitID != want.StartUnitID || got.EndUnitID != want.EndUnitID {
 			t.Fatalf("range[%d] = %#v, want %#v", i, result.Plan.Ranges[i], want)
 		}
+		if len(got.Annotations) != 0 {
+			t.Fatalf("range[%d] annotations = %#v, want absent", i, got.Annotations)
 		}
-	var firstAnnotation map[string]any
-	if err := json.Unmarshal(result.Plan.Ranges[0].Annotations[annotationNamespace], &firstAnnotation); err != nil {
-		t.Fatalf("decode first scene annotation: %v", err)
 	}
-	wantFirst := map[string]any{
-		"short_title": "Goblin parley", "primary_mode": "Discussion",
-		"main_participants":   []any{"Aria", "Goblin scout"},
-		"summary":             "The party negotiates with a scout.",
-		"boundary_note":       "The scene covers the discussion before fighting starts.",
-		"boundary_confidence": "High",
+	if len(result.Plan.Annotations) != 0 {
+		t.Fatalf("plan annotations = %#v, want absent", result.Plan.Annotations)
 	}
-	if !reflect.DeepEqual(firstAnnotation, wantFirst) {
-		t.Fatalf("first scene annotation = %#v, want %#v", firstAnnotation, wantFirst)
+	if len(result.Warnings) != 0 {
+		t.Fatalf("warnings = %#v, want absent", result.Warnings)
 	}
-	if len(firstAnnotation) != 6 {
-		t.Fatalf("first scene annotation keys = %#v, want exact six fields", firstAnnotation)
+}
+
+func TestPlanUsesDocumentOrderForNonconsecutiveUnitIDs(t *testing.T) {
+	doc := &source.SourceDocument{
+		ID:     "session-nonnumeric",
+		Kind:   "transcript",
+		Format: "application/vnd.seriatim.minimal+json",
+		Digest: "sha256:nonnumeric",
+		Units: []source.SourceUnit{
+			{ID: 10, Kind: "transcript_segment", Text: "The party arrives.", Ref: source.SourceRef{SourceID: "session-nonnumeric", StartUnitID: 10, EndUnitID: 10}},
+			{ID: 3, Kind: "transcript_segment", Text: "The party explores.", Ref: source.SourceRef{SourceID: "session-nonnumeric", StartUnitID: 3, EndUnitID: 3}},
+			{ID: 20, Kind: "transcript_segment", Text: "The party rests.", Ref: source.SourceRef{SourceID: "session-nonnumeric", StartUnitID: 20, EndUnitID: 20}},
+		},
 	}
-	var planAnnotation map[string]any
-	if err := json.Unmarshal(result.Plan.Annotations[annotationNamespace], &planAnnotation); err != nil {
-		t.Fatalf("decode plan annotation: %v", err)
+	req := chunkRequest()
+	req.Source = doc
+
+	t.Run("accepts document-ordered ranges", func(t *testing.T) {
+		client := &fakeScenesLLMClient{response: chunkResponse{Scenes: []sceneResponse{
+			scene(10, 3),
+			scene(20, 20),
+		}}}
+		result, err := newChunker(t, client).Plan(context.Background(), req)
+		if err != nil {
+			t.Fatalf("Plan() error = %v, want nil", err)
 		}
-	if !reflect.DeepEqual(planAnnotation, map[string]any{"boundary_caveats": []any{"The transition into combat is gradual."}}) || len(planAnnotation) != 1 {
-		t.Fatalf("plan annotation = %#v, want normalized caveats only", planAnnotation)
+		if got := result.Plan.Ranges; !reflect.DeepEqual(got, []source.ChunkRange{{StartUnitID: 10, EndUnitID: 3}, {StartUnitID: 20, EndUnitID: 20}}) {
+			t.Fatalf("ranges = %#v, want source-document order", got)
 		}
-	if got := result.Warnings; len(got) != 1 ||
-		got[0].Scope != Key ||
-		got[0].ReasonCode != "scene_boundary_caveat" ||
-		got[0].Message != "The transition into combat is gradual." {
-		t.Fatalf("Warnings = %#v, want boundary caveat warning", got)
+	})
+
+	t.Run("rejects reversed document positions", func(t *testing.T) {
+		client := &fakeScenesLLMClient{response: chunkResponse{Scenes: []sceneResponse{
+			scene(3, 10),
+			scene(20, 20),
+		}}}
+		_, err := newChunker(t, client).Plan(context.Background(), req)
+		if err == nil || !strings.Contains(err.Error(), "appears after") {
+			t.Fatalf("Plan() error = %v, want document-position reversal", err)
 		}
+	})
 }
 
 func TestPlanPassesReferencesAsPromptInputs(t *testing.T) {
 	client := &fakeScenesLLMClient{response: chunkResponse{
 		Scenes: []sceneResponse{
-			{
-				StartUnitID:        shared.UnitRefFromInt(1),
-				EndUnitID:          shared.UnitRefFromInt(4),
-				ShortTitle:         "Ambush",
-				PrimaryMode:        "Combat",
-				MainParticipants:   []string{"Aria"},
-				Summary:            "The party is ambushed.",
-				BoundaryNote:       "One scene covers the short fixture.",
-				BoundaryConfidence: "High",
-			},
+			scene(1, 4),
 		},
 	}}
 	req := chunkRequest()
@@ -298,51 +291,6 @@ func TestPromptInputsMapLegacyRosterReferenceToParty(t *testing.T) {
 	}
 }
 
-func TestPlanRejectsWhitespaceOnlyBoundaryCaveats(t *testing.T) {
-	client := &fakeScenesLLMClient{
-		response: chunkResponse{
-			Scenes: validSceneResponse().Scenes,
-			BoundaryCaveats: []string{
-				"   ",
-			},
-		},
-	}
-
-	_, err := newChunker(t, client).Plan(context.Background(), chunkRequest())
-	if err == nil {
-		t.Fatal("Plan() error = nil, want malformed structured output error")
-	}
-	if !strings.Contains(err.Error(), "dnd scenes chunker") || !strings.Contains(err.Error(), "malformed structured output") || !strings.Contains(err.Error(), "boundary_caveats[0]") {
-		t.Fatalf("Plan() error = %q, want malformed boundary caveat context", err.Error())
-	}
-}
-
-func TestPlanDefensivelyCopiesAnnotationValues(t *testing.T) {
-	doc := sceneSourceDocument()
-	client := &fakeScenesLLMClient{response: validSceneResponse()}
-
-	result, err := newChunker(t, client).Plan(context.Background(), contracts.ChunkRequest{
-		Source:      doc,
-		SourceInput: sceneSourceInput(),
-		SessionID:   "session-123",
-		LLMProfile:  "profile-scenes",
-	})
-	if err != nil {
-		t.Fatalf("Plan() error = %v, want nil", err)
-	}
-
-	client.response.Scenes[0].MainParticipants[0] = "mutated"
-	var annotation struct {
-		MainParticipants []string `json:"main_participants"`
-	}
-	if err := json.Unmarshal(result.Plan.Ranges[0].Annotations[annotationNamespace], &annotation); err != nil {
-		t.Fatalf("decode annotation: %v", err)
-	}
-	if !reflect.DeepEqual(annotation.MainParticipants, []string{"Aria"}) {
-		t.Fatalf("participants = %#v, want defensive copy", annotation.MainParticipants)
-	}
-}
-
 func TestChunkerManifestMetadataIncludesPromptAndSchemaProvenance(t *testing.T) {
 	metadata := newChunker(t, &fakeScenesLLMClient{}).ManifestMetadata()
 
@@ -455,38 +403,6 @@ func TestPlanRejectsMalformedStructuredOutput(t *testing.T) {
 			}),
 			want: "final scene",
 		},
-		{
-			name: "empty metadata field",
-			response: replaceScenes(validSceneResponse(), []sceneResponse{
-				{
-					StartUnitID:        shared.UnitRefFromInt(1),
-					EndUnitID:          shared.UnitRefFromInt(4),
-					ShortTitle:         " ",
-					PrimaryMode:        "Narrative",
-					MainParticipants:   []string{"Aria"},
-					Summary:            "Summary.",
-					BoundaryNote:       "Note.",
-					BoundaryConfidence: "High",
-				},
-			}),
-			want: "short_title",
-		},
-		{
-			name: "empty participant",
-			response: replaceScenes(validSceneResponse(), []sceneResponse{
-				{
-					StartUnitID:        shared.UnitRefFromInt(1),
-					EndUnitID:          shared.UnitRefFromInt(4),
-					ShortTitle:         "Title",
-					PrimaryMode:        "Narrative",
-					MainParticipants:   []string{"Aria", " "},
-					Summary:            "Summary.",
-					BoundaryNote:       "Note.",
-					BoundaryConfidence: "High",
-				},
-			}),
-			want: "main_participants",
-		},
 	}
 
 	for _, tt := range tests {
@@ -559,7 +475,6 @@ func validSceneResponse() chunkResponse {
 		Scenes: []sceneResponse{
 			scene(1, 4),
 		},
-		BoundaryCaveats: []string{},
 	}
 }
 
@@ -572,12 +487,6 @@ func scene(startUnitID int, endUnitID int) sceneResponse {
 	return sceneResponse{
 		StartUnitID: shared.UnitRefFromInt(startUnitID),
 		EndUnitID:   shared.UnitRefFromInt(endUnitID),
-		ShortTitle:         "Scene title",
-		PrimaryMode:        "Narrative",
-		MainParticipants:   []string{"Aria"},
-		Summary:            "A compact summary.",
-		BoundaryNote:       "The source units form one coherent scene.",
-		BoundaryConfidence: "High",
 	}
 }
 

internal/modules/dnd/chunk/scenes/model.go

@@ -4,27 +4,9 @@ import "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
 
 type chunkResponse struct {
 	Scenes []sceneResponse `json:"scenes"`
-	BoundaryCaveats []string        `json:"boundary_caveats"`
 }
 
 type sceneResponse struct {
 	StartUnitID shared.UnitRef `json:"start_unit_id"`
 	EndUnitID   shared.UnitRef `json:"end_unit_id"`
-	ShortTitle         string         `json:"short_title"`
-	PrimaryMode        string         `json:"primary_mode"`
-	MainParticipants   []string       `json:"main_participants"`
-	Summary            string         `json:"summary"`
-	BoundaryNote       string         `json:"boundary_note"`
-	BoundaryConfidence string         `json:"boundary_confidence"`
-}
-
-type normalizedScene struct {
-	StartUnitID        int
-	EndUnitID          int
-	ShortTitle         string
-	PrimaryMode        string
-	MainParticipants   []string
-	Summary            string
-	BoundaryNote       string
-	BoundaryConfidence string
 }

internal/modules/dnd/chunk/scenes/schema_test.go

@@ -35,7 +35,7 @@ func TestLoadResponseSchemaForScenes(t *testing.T) {
 	}
 }
 
-func TestResponseSchemaValidatesSceneResponses(t *testing.T) {
+func TestResponseSchemaValidatesMinimalSceneResponses(t *testing.T) {
 	schema, err := loadResponseSchema()
 	if err != nil {
 		t.Fatalf("loadResponseSchema() error = %v, want nil", err)
@@ -55,11 +55,15 @@ func TestResponseSchemaValidatesSceneResponses(t *testing.T) {
 		mutate func(map[string]any)
 	}{
 		{
-			name: "obsolete segment boundaries",
+			name: "missing scenes",
 			mutate: func(response map[string]any) {
-				scene := response["scenes"].([]any)[0].(map[string]any)
-				scene["start_segment_id"] = 1
-				scene["end_segment_id"] = 2
+				delete(response, "scenes")
+			},
+		},
+		{
+			name: "empty scenes",
+			mutate: func(response map[string]any) {
+				response["scenes"] = []any{}
 			},
 		},
 		{
@@ -75,27 +79,21 @@ func TestResponseSchemaValidatesSceneResponses(t *testing.T) {
 			},
 		},
 		{
-			name: "invalid primary mode",
+			name: "non-integer endpoint",
 			mutate: func(response map[string]any) {
-				response["scenes"].([]any)[0].(map[string]any)["primary_mode"] = "Unknown"
+				response["scenes"].([]any)[0].(map[string]any)["start_unit_id"] = 1.5
 			},
 		},
 		{
-			name: "invalid boundary confidence",
+			name: "unknown top-level field",
 			mutate: func(response map[string]any) {
-				response["scenes"].([]any)[0].(map[string]any)["boundary_confidence"] = "Unknown"
+				response["boundary_caveats"] = []any{}
 			},
 		},
 		{
-			name: "empty boundary caveat",
+			name: "unknown scene field",
 			mutate: func(response map[string]any) {
-				response["boundary_caveats"] = []any{""}
-			},
-		},
-		{
-			name: "unknown property",
-			mutate: func(response map[string]any) {
-				response["unexpected"] = true
+				response["scenes"].([]any)[0].(map[string]any)["short_title"] = "Old contract"
 			},
 		},
 	}
@@ -116,21 +114,7 @@ func TestResponseSchemaValidatesSceneResponses(t *testing.T) {
 }
 
 func TestResponseStructAcceptsIntegerBoundaries(t *testing.T) {
-	raw := []byte(`{
-		"scenes": [
-			{
-				"start_unit_id": 1,
-				"end_unit_id": 3,
-				"short_title": "Ambush",
-				"primary_mode": "Combat",
-				"main_participants": ["Aria"],
-				"summary": "The party fights.",
-				"boundary_note": "Combat starts and resolves.",
-				"boundary_confidence": "High"
-			}
-		],
-		"boundary_caveats": []
-	}`)
+	raw := []byte(`{"scenes":[{"start_unit_id":1,"end_unit_id":3}]}`)
 
 	var response chunkResponse
 	if err := json.Unmarshal(raw, &response); err != nil {
@@ -159,7 +143,7 @@ func TestResponseSchemaJSONIsMutationSafe(t *testing.T) {
 		t.Fatalf("schema JSON was mutated: %s", second.JSONSchema)
 	}
 	if len(second.JSONSchema) > 0 && second.JSONSchema[0] == '[' {
-		t.Fatalf("schema JSON did not use defensive copy")
+		t.Fatal("schema JSON did not use defensive copy")
 	}
 }
 
@@ -169,15 +153,8 @@ func validSceneSchemaResponse() map[string]any {
 			map[string]any{
 				"start_unit_id": 1,
 				"end_unit_id":   3,
-				"short_title":         "Ambush",
-				"primary_mode":        "Combat",
-				"main_participants":   []any{"Aria"},
-				"summary":             "The party fights.",
-				"boundary_note":       "Combat starts and resolves.",
-				"boundary_confidence": "High",
 			},
 		},
-		"boundary_caveats": []any{},
 	}
 }
 

internal/modules/dnd/codec/npcinteractions/assets/schemas/dnd_npc_interactions.v1.json

@@ -0,0 +1,50 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "notarius.dnd.npc_interactions",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["interactions"],
+  "properties": {
+    "interactions": {
+      "type": "array",
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "required": ["name", "kind", "source_refs"],
+        "properties": {
+          "name": {
+            "type": "string",
+            "minLength": 1
+          },
+          "kind": {
+            "type": "string",
+            "enum": ["mentioned", "noncombat_presence", "dialogue", "combat_ally", "combat_opponent", "other"]
+          },
+          "source_refs": {
+            "type": "array",
+            "minItems": 1,
+            "items": {
+              "type": "object",
+              "additionalProperties": false,
+              "required": ["source_id", "start_unit_id", "end_unit_id"],
+              "properties": {
+                "source_id": {
+                  "type": "string",
+                  "minLength": 1
+                },
+                "start_unit_id": {
+                  "type": "integer",
+                  "minimum": 1
+                },
+                "end_unit_id": {
+                  "type": "integer",
+                  "minimum": 1
+                }
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+}

internal/modules/dnd/codec/npcinteractions/codec.go

@@ -0,0 +1,140 @@
+// Package npcinteractions encodes durable D&D NPC interaction artifacts.
+package npcinteractions
+
+import (
+	"bytes"
+	"embed"
+	"encoding/json"
+	"fmt"
+	"io"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+const (
+	SchemaID      = "notarius.dnd.npc_interactions"
+	SchemaName    = "notarius_dnd_npc_interactions_v1"
+	SchemaVersion = "v1"
+	MediaType     = "application/json"
+)
+
+//go:embed assets/schemas/dnd_npc_interactions.v1.json
+var schemaAssets embed.FS
+
+var _ contracts.ArtifactCodec[dnd.NPCInteractionList] = (*Codec)(nil)
+
+type Codec struct{}
+
+func New() *Codec { return &Codec{} }
+
+func (c *Codec) Kind() contracts.ArtifactKind { return dnd.NPCInteractionListKind }
+
+func (c *Codec) Schema() contracts.ArtifactSchema {
+	raw, err := schemaAssets.ReadFile("assets/schemas/dnd_npc_interactions.v1.json")
+	if err != nil {
+		return contracts.ArtifactSchema{}
+	}
+	return contracts.ArtifactSchema{
+		ID:         SchemaID,
+		Name:       SchemaName,
+		Version:    SchemaVersion,
+		JSONSchema: append([]byte(nil), raw...),
+	}
+}
+
+func (c *Codec) MediaType() string { return MediaType }
+
+func (c *Codec) Metadata(value dnd.NPCInteractionList) map[string]any {
+	return map[string]any{"interaction_count": len(value.Interactions)}
+}
+
+func (c *Codec) Encode(value dnd.NPCInteractionList) ([]byte, error) {
+	if err := validate(value); err != nil {
+		return nil, fmt.Errorf("encode dnd npc interaction list: %w", err)
+	}
+	return c.EncodeCandidate(value)
+}
+
+// EncodeCandidate provides the durable representation before semantic
+// validators have approved a value.
+func (c *Codec) EncodeCandidate(value dnd.NPCInteractionList) ([]byte, error) {
+	content, err := json.Marshal(value)
+	if err != nil {
+		return nil, fmt.Errorf("encode dnd npc interaction list: %w", err)
+	}
+	return content, nil
+}
+
+func (c *Codec) Decode(content []byte) (dnd.NPCInteractionList, error) {
+	value, err := c.DecodeCandidate(content)
+	if err != nil {
+		return dnd.NPCInteractionList{}, err
+	}
+	if err := validate(value); err != nil {
+		return dnd.NPCInteractionList{}, fmt.Errorf("decode dnd npc interaction list: %w", err)
+	}
+	return value, nil
+}
+
+// DecodeCandidate reads one strict durable JSON value before semantic
+// validators have approved it.
+func (c *Codec) DecodeCandidate(content []byte) (dnd.NPCInteractionList, error) {
+	decoder := json.NewDecoder(bytes.NewReader(content))
+	decoder.DisallowUnknownFields()
+	var value dnd.NPCInteractionList
+	if err := decoder.Decode(&value); err != nil {
+		return dnd.NPCInteractionList{}, fmt.Errorf("decode dnd npc interaction list: %w", err)
+	}
+	var trailing any
+	if err := decoder.Decode(&trailing); err != io.EOF {
+		return dnd.NPCInteractionList{}, fmt.Errorf("decode dnd npc interaction list: multiple JSON values")
+	}
+	return value, nil
+}
+
+func validate(value dnd.NPCInteractionList) error {
+	if value.Interactions == nil {
+		return fmt.Errorf("interactions must be present")
+	}
+	for index, interaction := range value.Interactions {
+		prefix := fmt.Sprintf("interactions[%d]", index)
+		if strings.TrimSpace(interaction.Name) == "" {
+			return fmt.Errorf("%s.name must not be empty", prefix)
+		}
+		if !validInteractionKind(interaction.Kind) {
+			return fmt.Errorf("%s.kind must be supported", prefix)
+		}
+		if len(interaction.SourceRefs) == 0 {
+			return fmt.Errorf("%s.source_refs must contain at least one reference", prefix)
+		}
+		for refIndex, ref := range interaction.SourceRefs {
+			refPrefix := fmt.Sprintf("%s.source_refs[%d]", prefix, refIndex)
+			if strings.TrimSpace(ref.SourceID) == "" {
+				return fmt.Errorf("%s.source_id must not be empty", refPrefix)
+			}
+			if ref.StartUnitID <= 0 {
+				return fmt.Errorf("%s.start_unit_id must be positive", refPrefix)
+			}
+			if ref.EndUnitID <= 0 {
+				return fmt.Errorf("%s.end_unit_id must be positive", refPrefix)
+			}
+		}
+	}
+	return nil
+}
+
+func validInteractionKind(value dnd.NPCInteractionKind) bool {
+	switch value {
+	case dnd.NPCInteractionKindMentioned,
+		dnd.NPCInteractionKindNoncombatPresence,
+		dnd.NPCInteractionKindDialogue,
+		dnd.NPCInteractionKindCombatAlly,
+		dnd.NPCInteractionKindCombatOpponent,
+		dnd.NPCInteractionKindOther:
+		return true
+	default:
+		return false
+	}
+}

internal/modules/dnd/codec/npcinteractions/codec_test.go

@@ -0,0 +1,183 @@
+package npcinteractions
+
+import (
+	"bytes"
+	"encoding/json"
+	"errors"
+	"os"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+func validList() dnd.NPCInteractionList {
+	return dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{
+		{
+			Name: "Mira Thorn", Kind: dnd.NPCInteractionKindDialogue,
+			SourceRefs: []source.SourceRef{{SourceID: "session-alpha", StartUnitID: 1, EndUnitID: 2}},
+		},
+		{
+			Name: "Hooded Guard", Kind: dnd.NPCInteractionKindCombatOpponent,
+			SourceRefs: []source.SourceRef{{SourceID: "session-alpha", StartUnitID: 3, EndUnitID: 3}},
+		},
+	}}
+}
+
+func TestCodecMatchesMaintainedDurableFixture(t *testing.T) {
+	raw, err := os.ReadFile("testdata/dnd_npc_interactions.v1.json")
+	if err != nil {
+		t.Fatal(err)
+	}
+	codec := New()
+	value, err := codec.Decode(raw)
+	if err != nil {
+		t.Fatalf("Decode() error = %v", err)
+	}
+	if want := validList(); !reflect.DeepEqual(value, want) {
+		t.Fatalf("Decode() = %#v, want %#v", value, want)
+	}
+	encoded, err := codec.Encode(value)
+	if err != nil {
+		t.Fatalf("Encode() error = %v", err)
+	}
+	var compact bytes.Buffer
+	if err := json.Compact(&compact, raw); err != nil {
+		t.Fatal(err)
+	}
+	if !bytes.Equal(encoded, compact.Bytes()) {
+		t.Fatalf("Encode() = %s, want %s", encoded, compact.Bytes())
+	}
+}
+
+func TestCodecOwnsDurableSchemaAndRegistersExactType(t *testing.T) {
+	codec := New()
+	schema := codec.Schema()
+	if codec.Kind() != dnd.NPCInteractionListKind || codec.MediaType() != MediaType {
+		t.Fatalf("codec identity = %q/%q", codec.Kind(), codec.MediaType())
+	}
+	if schema.ID != SchemaID || schema.Name != SchemaName || schema.Version != SchemaVersion || !json.Valid(schema.JSONSchema) {
+		t.Fatalf("schema = %#v", schema)
+	}
+	registry := pipeline.NewArtifactCodecRegistry()
+	if err := pipeline.RegisterArtifactCodec(registry, codec); err != nil {
+		t.Fatal(err)
+	}
+	spec, ok := registry.Spec(dnd.NPCInteractionListKind)
+	if !ok || spec.SchemaDigest != contracts.DigestArtifactSchema(schema) {
+		t.Fatalf("registered spec = %#v, %t", spec, ok)
+	}
+	if _, err := registry.Encode(dnd.NPCInteractionListKind, dnd.NPCList{}); err == nil {
+		t.Fatal("Encode() error = nil, want exact type rejection")
+	} else {
+		var typeErr *pipeline.ArtifactCodecTypeError
+		if !errors.As(err, &typeErr) {
+			t.Fatalf("Encode() error = %T, want ArtifactCodecTypeError", err)
+		}
+	}
+}
+
+func TestCodecSupportsEmptyListAndPreservesCollectionPresenceInCandidates(t *testing.T) {
+	codec := New()
+	empty := dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{}}
+	content, err := codec.Encode(empty)
+	if err != nil || string(content) != `{"interactions":[]}` {
+		t.Fatalf("Encode() = %s, %v", content, err)
+	}
+	for _, candidate := range []dnd.NPCInteractionList{
+		{},
+		empty,
+		{Interactions: []dnd.NPCInteraction{{Name: " ", Kind: "unsupported", SourceRefs: nil}}},
+		{Interactions: []dnd.NPCInteraction{{Name: " ", Kind: "unsupported", SourceRefs: []source.SourceRef{}}}},
+		{Interactions: []dnd.NPCInteraction{{Name: " ", Kind: "unsupported", SourceRefs: []source.SourceRef{{StartUnitID: 0, EndUnitID: -1}}}}},
+	} {
+		content, err := codec.EncodeCandidate(candidate)
+		if err != nil || !json.Valid(content) {
+			t.Fatalf("EncodeCandidate() = %s, %v", content, err)
+		}
+		decoded, err := codec.DecodeCandidate(content)
+		if err != nil || !reflect.DeepEqual(decoded, candidate) {
+			t.Fatalf("DecodeCandidate() = %#v, %v; want %#v", decoded, err, candidate)
+		}
+	}
+}
+
+func TestCodecStrictlyRejectsMalformedUnknownAndTrailingJSON(t *testing.T) {
+	validJSON := `{"interactions":[{"name":"Mira Thorn","kind":"dialogue","source_refs":[{"source_id":"session","start_unit_id":1,"end_unit_id":1}]}]}`
+	for _, test := range []struct{ name, raw, want string }{
+		{"malformed", `{`, "decode dnd npc interaction list"},
+		{"unknown top-level", `{"interactions":[],"unexpected":true}`, "unknown field"},
+		{"unknown interaction field", strings.Replace(validJSON, `"kind":"dialogue"`, `"kind":"dialogue","unexpected":true`, 1), "unknown field"},
+		{"unknown source reference field", strings.Replace(validJSON, `"end_unit_id":1`, `"end_unit_id":1,"unexpected":true`, 1), "unknown field"},
+		{"trailing", `{"interactions":[]} {}`, "multiple JSON values"},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := New().Decode([]byte(test.raw)); err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("Decode() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+}
+
+func TestCodecRejectsRequiredShapeEnumAndReferenceBoundaries(t *testing.T) {
+	tests := []struct {
+		name  string
+		value dnd.NPCInteractionList
+		want  string
+	}{
+		{"nil interactions", dnd.NPCInteractionList{}, "interactions must be present"},
+		{"blank name", mutate(validList(), func(v *dnd.NPCInteractionList) { v.Interactions[0].Name = " " }), "name must not be empty"},
+		{"unsupported kind", mutate(validList(), func(v *dnd.NPCInteractionList) { v.Interactions[0].Kind = "unsupported" }), "kind must be supported"},
+		{"nil source refs", mutate(validList(), func(v *dnd.NPCInteractionList) { v.Interactions[0].SourceRefs = nil }), "source_refs must contain"},
+		{"empty source ID", mutate(validList(), func(v *dnd.NPCInteractionList) { v.Interactions[0].SourceRefs[0].SourceID = " " }), "source_id must not be empty"},
+		{"non-positive start", mutate(validList(), func(v *dnd.NPCInteractionList) { v.Interactions[0].SourceRefs[0].StartUnitID = 0 }), "start_unit_id must be positive"},
+		{"non-positive end", mutate(validList(), func(v *dnd.NPCInteractionList) { v.Interactions[0].SourceRefs[0].EndUnitID = 0 }), "end_unit_id must be positive"},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := New().Encode(test.value); err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("Encode() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+}
+
+func TestCodecAcceptsEveryInteractionKind(t *testing.T) {
+	for _, kind := range []dnd.NPCInteractionKind{
+		dnd.NPCInteractionKindMentioned,
+		dnd.NPCInteractionKindNoncombatPresence,
+		dnd.NPCInteractionKindDialogue,
+		dnd.NPCInteractionKindCombatAlly,
+		dnd.NPCInteractionKindCombatOpponent,
+		dnd.NPCInteractionKindOther,
+	} {
+		value := validList()
+		value.Interactions[0].Kind = kind
+		if _, err := New().Encode(value); err != nil {
+			t.Fatalf("Encode(%q) error = %v", kind, err)
+		}
+	}
+}
+
+func TestCodecSchemaAndMetadataAreDefensive(t *testing.T) {
+	codec := New()
+	first := codec.Schema()
+	first.JSONSchema[0] = '['
+	if second := codec.Schema(); !json.Valid(second.JSONSchema) || second.JSONSchema[0] == '[' {
+		t.Fatal("Schema() returned shared bytes")
+	}
+	metadata := codec.Metadata(validList())
+	metadata["other"] = true
+	if next := codec.Metadata(validList()); len(next) != 1 || next["interaction_count"] != 2 {
+		t.Fatalf("Metadata() = %#v", next)
+	}
+}
+
+func mutate(value dnd.NPCInteractionList, change func(*dnd.NPCInteractionList)) dnd.NPCInteractionList {
+	change(&value)
+	return value
+}

internal/modules/dnd/codec/npcinteractions/testdata/dnd_npc_interactions.v1.json

@@ -0,0 +1,18 @@
+{
+  "interactions": [
+    {
+      "name": "Mira Thorn",
+      "kind": "dialogue",
+      "source_refs": [
+        {"source_id": "session-alpha", "start_unit_id": 1, "end_unit_id": 2}
+      ]
+    },
+    {
+      "name": "Hooded Guard",
+      "kind": "combat_opponent",
+      "source_refs": [
+        {"source_id": "session-alpha", "start_unit_id": 3, "end_unit_id": 3}
+      ]
+    }
+  ]
+}

internal/modules/dnd/codec/scenedescriptions/assets/schemas/dnd_scene_descriptions.v1.json

@@ -0,0 +1,55 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "notarius.dnd.scene_descriptions",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["scenes"],
+  "properties": {
+    "scenes": {
+      "type": "array",
+      "minItems": 1,
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "required": ["id", "source_ref", "kind", "title", "summary"],
+        "properties": {
+          "id": {
+            "type": "string",
+            "minLength": 1
+          },
+          "source_ref": {
+            "type": "object",
+            "additionalProperties": false,
+            "required": ["source_id", "start_unit_id", "end_unit_id"],
+            "properties": {
+              "source_id": {
+                "type": "string",
+                "minLength": 1
+              },
+              "start_unit_id": {
+                "type": "integer",
+                "minimum": 1
+              },
+              "end_unit_id": {
+                "type": "integer",
+                "minimum": 1
+              }
+            }
+          },
+          "kind": {
+            "type": "string",
+            "enum": ["combat", "narrative", "recap", "meta"]
+          },
+          "title": {
+            "type": "string",
+            "minLength": 1
+          },
+          "summary": {
+            "type": "string",
+            "minLength": 1
+          }
+        }
+      }
+    }
+  }
+}

internal/modules/dnd/codec/scenedescriptions/codec.go

@@ -0,0 +1,137 @@
+package scenedescriptions
+
+import (
+	"bytes"
+	"embed"
+	"encoding/json"
+	"fmt"
+	"io"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+const (
+	SchemaID      = "notarius.dnd.scene_descriptions"
+	SchemaName    = "notarius_dnd_scene_descriptions_v1"
+	SchemaVersion = "v1"
+	MediaType     = "application/json"
+)
+
+//go:embed assets/schemas/dnd_scene_descriptions.v1.json
+var schemaAssets embed.FS
+
+var _ contracts.ArtifactCodec[dnd.SceneDescriptionList] = (*Codec)(nil)
+
+type Codec struct{}
+
+func New() *Codec { return &Codec{} }
+
+func (c *Codec) Kind() contracts.ArtifactKind { return dnd.SceneDescriptionListKind }
+
+func (c *Codec) Schema() contracts.ArtifactSchema {
+	raw, err := schemaAssets.ReadFile("assets/schemas/dnd_scene_descriptions.v1.json")
+	if err != nil {
+		return contracts.ArtifactSchema{}
+	}
+	return contracts.ArtifactSchema{
+		ID:         SchemaID,
+		Name:       SchemaName,
+		Version:    SchemaVersion,
+		JSONSchema: append([]byte(nil), raw...),
+	}
+}
+
+func (c *Codec) MediaType() string { return MediaType }
+
+func (c *Codec) Metadata(value dnd.SceneDescriptionList) map[string]any {
+	return map[string]any{"scene_count": len(value.Scenes)}
+}
+
+func (c *Codec) Encode(value dnd.SceneDescriptionList) ([]byte, error) {
+	if err := validate(value); err != nil {
+		return nil, fmt.Errorf("encode dnd scene description list: %w", err)
+	}
+	return c.EncodeCandidate(value)
+}
+
+// EncodeCandidate provides the durable representation before typed validators
+// have approved a value.
+func (c *Codec) EncodeCandidate(value dnd.SceneDescriptionList) ([]byte, error) {
+	content, err := json.Marshal(value)
+	if err != nil {
+		return nil, fmt.Errorf("encode dnd scene description list: %w", err)
+	}
+	return content, nil
+}
+
+func (c *Codec) Decode(content []byte) (dnd.SceneDescriptionList, error) {
+	value, err := c.DecodeCandidate(content)
+	if err != nil {
+		return dnd.SceneDescriptionList{}, err
+	}
+	if err := validate(value); err != nil {
+		return dnd.SceneDescriptionList{}, fmt.Errorf("decode dnd scene description list: %w", err)
+	}
+	return value, nil
+}
+
+// DecodeCandidate reads one strict durable JSON value before typed validators
+// have approved it.
+func (c *Codec) DecodeCandidate(content []byte) (dnd.SceneDescriptionList, error) {
+	decoder := json.NewDecoder(bytes.NewReader(content))
+	decoder.DisallowUnknownFields()
+	var value dnd.SceneDescriptionList
+	if err := decoder.Decode(&value); err != nil {
+		return dnd.SceneDescriptionList{}, fmt.Errorf("decode dnd scene description list: %w", err)
+	}
+	var trailing any
+	if err := decoder.Decode(&trailing); err != io.EOF {
+		return dnd.SceneDescriptionList{}, fmt.Errorf("decode dnd scene description list: multiple JSON values")
+	}
+	return value, nil
+}
+
+func validate(value dnd.SceneDescriptionList) error {
+	if value.Scenes == nil {
+		return fmt.Errorf("scenes must be present")
+	}
+	if len(value.Scenes) == 0 {
+		return fmt.Errorf("scenes must not be empty")
+	}
+	for index, scene := range value.Scenes {
+		prefix := fmt.Sprintf("scenes[%d]", index)
+		if strings.TrimSpace(scene.ID) == "" {
+			return fmt.Errorf("%s.id must not be empty", prefix)
+		}
+		if !validSceneKind(scene.Kind) {
+			return fmt.Errorf("%s.kind must be supported", prefix)
+		}
+		if strings.TrimSpace(scene.Title) == "" {
+			return fmt.Errorf("%s.title must not be empty", prefix)
+		}
+		if strings.TrimSpace(scene.Summary) == "" {
+			return fmt.Errorf("%s.summary must not be empty", prefix)
+		}
+		if strings.TrimSpace(scene.SourceRef.SourceID) == "" {
+			return fmt.Errorf("%s.source_ref.source_id must not be empty", prefix)
+		}
+		if scene.SourceRef.StartUnitID <= 0 {
+			return fmt.Errorf("%s.source_ref.start_unit_id must be positive", prefix)
+		}
+		if scene.SourceRef.EndUnitID <= 0 {
+			return fmt.Errorf("%s.source_ref.end_unit_id must be positive", prefix)
+		}
+	}
+	return nil
+}
+
+func validSceneKind(value dnd.SceneKind) bool {
+	switch value {
+	case dnd.SceneKindCombat, dnd.SceneKindNarrative, dnd.SceneKindRecap, dnd.SceneKindMeta:
+		return true
+	default:
+		return false
+	}
+}

internal/modules/dnd/codec/scenedescriptions/codec_test.go

@@ -0,0 +1,208 @@
+package scenedescriptions
+
+import (
+	"bytes"
+	"encoding/json"
+	"os"
+	"reflect"
+	"strings"
+	"testing"
+
+	"github.com/santhosh-tekuri/jsonschema/v6"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+func validList() dnd.SceneDescriptionList {
+	return dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{
+		ID:        "chunk-000001",
+		SourceRef: source.SourceRef{SourceID: "session-alpha", StartUnitID: 1, EndUnitID: 2},
+		Kind:      dnd.SceneKindNarrative,
+		Title:     "At the city gate",
+		Summary:   "The party enters the city after speaking with its guard.",
+	}}}
+}
+
+func TestCodecMatchesMaintainedDurableFixture(t *testing.T) {
+	raw, err := os.ReadFile("testdata/dnd_scene_descriptions.v1.json")
+	if err != nil {
+		t.Fatal(err)
+	}
+	codec := New()
+	value, err := codec.Decode(raw)
+	if err != nil {
+		t.Fatalf("Decode() error = %v", err)
+	}
+	if want := validList(); !reflect.DeepEqual(value, want) {
+		t.Fatalf("Decode() = %#v, want %#v", value, want)
+	}
+	encoded, err := codec.Encode(value)
+	if err != nil {
+		t.Fatalf("Encode() error = %v", err)
+	}
+	var compact bytes.Buffer
+	if err := json.Compact(&compact, raw); err != nil {
+		t.Fatal(err)
+	}
+	if !bytes.Equal(encoded, compact.Bytes()) {
+		t.Fatalf("Encode() = %s, want %s", encoded, compact.Bytes())
+	}
+}
+
+func TestCodecOwnsDurableSchemaAndMetadata(t *testing.T) {
+	codec := New()
+	schema := codec.Schema()
+	if codec.Kind() != dnd.SceneDescriptionListKind || codec.MediaType() != MediaType {
+		t.Fatalf("codec identity = %q/%q", codec.Kind(), codec.MediaType())
+	}
+	if schema.ID != SchemaID || schema.Name != SchemaName || schema.Version != SchemaVersion || !json.Valid(schema.JSONSchema) {
+		t.Fatalf("schema = %#v", schema)
+	}
+	registry := pipeline.NewArtifactCodecRegistry()
+	if err := pipeline.RegisterArtifactCodec(registry, codec); err != nil {
+		t.Fatal(err)
+	}
+	spec, ok := registry.Spec(dnd.SceneDescriptionListKind)
+	if !ok || spec.SchemaDigest != contracts.DigestArtifactSchema(schema) {
+		t.Fatalf("registered spec = %#v, %t", spec, ok)
+	}
+	schema.JSONSchema[0] = '['
+	if next := codec.Schema(); !json.Valid(next.JSONSchema) || next.JSONSchema[0] == '[' {
+		t.Fatal("Schema() returned shared bytes")
+	}
+	metadata := codec.Metadata(validList())
+	metadata["other"] = true
+	if next := codec.Metadata(validList()); len(next) != 1 || next["scene_count"] != 1 {
+		t.Fatalf("Metadata() = %#v", next)
+	}
+	encoded, err := codec.Encode(validList())
+	if err != nil {
+		t.Fatal(err)
+	}
+	encoded[0] = '['
+	if next, err := codec.Encode(validList()); err != nil || !json.Valid(next) || next[0] == '[' {
+		t.Fatalf("Encode() returned shared bytes: %s, %v", next, err)
+	}
+}
+
+func TestDurableSchemaRequiresAtLeastOneScene(t *testing.T) {
+	schemaDocument, err := jsonschema.UnmarshalJSON(bytes.NewReader(New().Schema().JSONSchema))
+	if err != nil {
+		t.Fatal(err)
+	}
+	compiler := jsonschema.NewCompiler()
+	if err := compiler.AddResource("schema.json", schemaDocument); err != nil {
+		t.Fatal(err)
+	}
+	compiled, err := compiler.Compile("schema.json")
+	if err != nil {
+		t.Fatal(err)
+	}
+	for _, test := range []struct {
+		name    string
+		content []byte
+		valid   bool
+	}{
+		{name: "one scene", content: mustEncode(t, validList()), valid: true},
+		{name: "empty scenes", content: []byte(`{"scenes":[]}`)},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			instance, err := jsonschema.UnmarshalJSON(bytes.NewReader(test.content))
+			if err != nil {
+				t.Fatal(err)
+			}
+			err = compiled.Validate(instance)
+			if (err == nil) != test.valid {
+				t.Fatalf("Validate() error = %v, want valid=%t", err, test.valid)
+			}
+		})
+	}
+}
+
+func TestCodecStrictlyRejectsInvalidDurableBoundaries(t *testing.T) {
+	validJSON := `{"scenes":[{"id":"chunk-000001","source_ref":{"source_id":"session","start_unit_id":1,"end_unit_id":1},"kind":"narrative","title":"Arrival","summary":"The party arrives."}]}`
+	tests := []struct{ name, raw, want string }{
+		{"malformed", `{`, "decode dnd scene description list"},
+		{"unknown top-level", `{"scenes":[],"unexpected":true}`, "unknown field"},
+		{"unknown scene field", strings.Replace(validJSON, `"kind":"narrative"`, `"kind":"narrative","unexpected":true`, 1), "unknown field"},
+		{"unknown source field", strings.Replace(validJSON, `"end_unit_id":1`, `"end_unit_id":1,"unexpected":true`, 1), "unknown field"},
+		{"trailing", `{"scenes":[]} {}`, "multiple JSON values"},
+		{"missing scenes", `{}`, "scenes must be present"},
+		{"empty scenes", `{"scenes":[]}`, "scenes must not be empty"},
+		{"invalid kind", strings.Replace(validJSON, `"kind":"narrative"`, `"kind":"other"`, 1), "kind must be supported"},
+		{"blank title", strings.Replace(validJSON, `"title":"Arrival"`, `"title":" "`, 1), "title must not be empty"},
+		{"blank summary", strings.Replace(validJSON, `"summary":"The party arrives."`, `"summary":" "`, 1), "summary must not be empty"},
+		{"blank ID", strings.Replace(validJSON, `"id":"chunk-000001"`, `"id":" "`, 1), "id must not be empty"},
+		{"blank source ID", strings.Replace(validJSON, `"source_id":"session"`, `"source_id":" "`, 1), "source_id must not be empty"},
+		{"invalid source start", strings.Replace(validJSON, `"start_unit_id":1`, `"start_unit_id":0`, 1), "start_unit_id must be positive"},
+		{"invalid source end", strings.Replace(validJSON, `"end_unit_id":1`, `"end_unit_id":0`, 1), "end_unit_id must be positive"},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := New().Decode([]byte(test.raw)); err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("Decode() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+}
+
+func TestCodecCandidatePreservesValidatorOwnedValues(t *testing.T) {
+	candidates := []dnd.SceneDescriptionList{
+		{},
+		{Scenes: []dnd.SceneDescription{}},
+		{Scenes: []dnd.SceneDescription{{ID: " ", Kind: "unsupported", Title: " ", Summary: " ", SourceRef: source.SourceRef{}}}},
+		{Scenes: []dnd.SceneDescription{{ID: " ", Kind: "unsupported", Title: " ", Summary: " ", SourceRef: source.SourceRef{StartUnitID: 0, EndUnitID: -1}}}},
+	}
+	for _, candidate := range candidates {
+		content, err := New().EncodeCandidate(candidate)
+		if err != nil || !json.Valid(content) {
+			t.Fatalf("EncodeCandidate() = %s, %v", content, err)
+		}
+		decoded, err := New().DecodeCandidate(content)
+		if err != nil || !reflect.DeepEqual(decoded, candidate) {
+			t.Fatalf("DecodeCandidate() = %#v, %v; want %#v", decoded, err, candidate)
+		}
+	}
+}
+
+func TestCodecRejectsRequiredApprovedValues(t *testing.T) {
+	tests := []struct {
+		name  string
+		value dnd.SceneDescriptionList
+		want  string
+	}{
+		{"nil scenes", dnd.SceneDescriptionList{}, "scenes must be present"},
+		{"empty scenes", dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{}}, "scenes must not be empty"},
+		{"empty ID", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].ID = " " }), "id must not be empty"},
+		{"unsupported kind", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].Kind = "unsupported" }), "kind must be supported"},
+		{"empty title", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].Title = " " }), "title must not be empty"},
+		{"empty summary", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].Summary = " " }), "summary must not be empty"},
+		{"empty source ID", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].SourceRef.SourceID = " " }), "source_id must not be empty"},
+		{"non-positive start", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].SourceRef.StartUnitID = 0 }), "start_unit_id must be positive"},
+		{"non-positive end", mutate(validList(), func(value *dnd.SceneDescriptionList) { value.Scenes[0].SourceRef.EndUnitID = 0 }), "end_unit_id must be positive"},
+	}
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := New().Encode(test.value); err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("Encode() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+}
+
+func mustEncode(t *testing.T, value dnd.SceneDescriptionList) []byte {
+	t.Helper()
+	content, err := json.Marshal(value)
+	if err != nil {
+		t.Fatal(err)
+	}
+	return content
+}
+
+func mutate(value dnd.SceneDescriptionList, change func(*dnd.SceneDescriptionList)) dnd.SceneDescriptionList {
+	change(&value)
+	return value
+}

internal/modules/dnd/codec/scenedescriptions/testdata/dnd_scene_descriptions.v1.json

@@ -0,0 +1 @@
+{"scenes":[{"id":"chunk-000001","source_ref":{"source_id":"session-alpha","start_unit_id":1,"end_unit_id":2},"kind":"narrative","title":"At the city gate","summary":"The party enters the city after speaking with its guard."}]}

internal/modules/dnd/extract/npcinteractions/assets.go

@@ -0,0 +1,6 @@
+package npcinteractions
+
+import "embed"
+
+//go:embed assets/schemas/dnd_npc_interactions_llm.v1.json assets/prompts/*.yaml assets/prompts/*.md
+var embeddedAssets embed.FS

internal/modules/dnd/extract/npcinteractions/assets/prompts/dnd.npc_interactions.yaml

@@ -0,0 +1,49 @@
+id: dnd.npc_interactions
+version: "v1"
+default_profile: gemini-2-flash
+inputs:
+  - name: transcript
+    required: true
+    content_type: application/json
+  - name: players
+    required: false
+    content_type: text/plain
+  - name: party
+    required: false
+    content_type: text/plain
+  - name: glossary
+    required: false
+    content_type: text/plain
+  - name: npcs
+    required: true
+    content_type: application/json
+messages:
+  - role: system
+    content_file: ./sharedassets/common-dnd-system.md
+  - role: user
+    content_file: ./sharedassets/common-dnd-extraction-evidence.md
+  - role: user
+    content_file: ./sharedassets/common-dnd-identity.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./sharedassets/common-dnd-references.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./sharedassets/common-dnd-npcs.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./task.md
+  - role: user
+    content_file: ./instructions.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./sharedassets/common-dnd-transcript.md
+output:
+  format: json
+  validation_mode: json_schema
+  schema_path: dnd_npc_interactions_llm.v1.json
+  repair_attempts: 0

internal/modules/dnd/extract/npcinteractions/assets/prompts/instructions.md

@@ -0,0 +1,18 @@
+Return the interactions array even when no interaction is established. Every
+record must contain name, kind, and source_refs. Cite transcript ranges that
+support both the NPC identity and the interaction kind.
+
+Use exactly one kind per occurrence:
+
+- mentioned: the NPC is referred to but is not established as present or communicating;
+- noncombat_presence: the NPC is present and relevant but does not meaningfully participate in dialogue or combat;
+- dialogue: the NPC speaks, responds, or is directly engaged in a meaningful non-combat exchange;
+- combat_ally: the NPC actively participates in combat on the party's side;
+- combat_opponent: the NPC actively participates in combat against the party; or
+- other: the transcript clearly establishes a direct NPC occurrence that fits none of the preceding kinds.
+
+When activities overlap, active combat participation outranks dialogue,
+presence, and mention; dialogue outranks noncombat presence and mention; and
+noncombat presence outranks mention. Other is only for directly evidenced
+activity outside those categories. Split an occurrence rather than assigning
+both combat alignments.

internal/modules/dnd/extract/npcinteractions/assets/prompts/task.md

@@ -0,0 +1,16 @@
+Extract Dungeons & Dragons NPC interaction occurrences from the supplied
+transcript.
+
+Include an occurrence only when the transcript establishes one supplied NPC,
+one interaction kind, and a coherent passage supporting both. Use only names
+from the supplied NPC registry. The registry helps ground identity but never
+proves that an interaction occurred.
+
+Do not summarize, infer relationships, sentiment, factions, motives, aliases,
+or persistent state. Do not identify player characters, anonymous groups, or
+invented NPCs. Return an empty interactions array when no supplied NPC has an
+evidenced interaction in this transcript passage.
+
+Keep occurrences within this transcript chunk. Split records when an NPC's
+interaction kind changes, when combat alignment changes, or when an NPC is
+first mentioned and later becomes present.

internal/modules/dnd/extract/npcinteractions/assets/schemas/dnd_npc_interactions_llm.v1.json

@@ -0,0 +1,41 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "notarius.dnd.npc_interactions.llm",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["interactions"],
+  "properties": {
+    "interactions": {
+      "type": "array",
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "required": ["name", "kind", "source_refs"],
+        "properties": {
+          "name": {
+            "type": "string"
+          },
+          "kind": {
+            "type": "string"
+          },
+          "source_refs": {
+            "type": "array",
+            "items": {
+              "type": "object",
+              "additionalProperties": false,
+              "required": ["start_unit_id", "end_unit_id"],
+              "properties": {
+                "start_unit_id": {
+                  "type": "integer"
+                },
+                "end_unit_id": {
+                  "type": "integer"
+                }
+              }
+            }
+          }
+        }
+      }
+    }
+  }
+}

internal/modules/dnd/extract/npcinteractions/canonicalize.go

@@ -0,0 +1,113 @@
+package npcinteractions
+
+import (
+	"sort"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+func canonicalizeResponse(response *extractionResponse, doc *source.SourceDocument) {
+	if response == nil {
+		return
+	}
+	for index := range response.Interactions {
+		canonicalizeInteraction(&response.Interactions[index])
+	}
+	sort.SliceStable(response.Interactions, func(i, j int) bool {
+		left, leftOK := earliestSourcePosition(doc, response.Interactions[i])
+		right, rightOK := earliestSourcePosition(doc, response.Interactions[j])
+		if leftOK != rightOK {
+			return leftOK
+		}
+		if !leftOK {
+			return false
+		}
+		return left < right
+	})
+}
+
+func canonicalizeInteraction(interaction *interactionResponse) {
+	if interaction == nil {
+		return
+	}
+	sort.SliceStable(interaction.SourceRefs, func(i, j int) bool {
+		left := interaction.SourceRefs[i]
+		right := interaction.SourceRefs[j]
+		if unitSortValue(left.StartUnitID) != unitSortValue(right.StartUnitID) {
+			return unitSortValue(left.StartUnitID) < unitSortValue(right.StartUnitID)
+		}
+		return unitSortValue(left.EndUnitID) < unitSortValue(right.EndUnitID)
+	})
+	interaction.SourceRefs = dedupeSourceRefs(interaction.SourceRefs)
+}
+
+func dedupeSourceRefs(refs []interactionSourceRefResponse) []interactionSourceRefResponse {
+	if len(refs) < 2 {
+		return refs
+	}
+	out := refs[:0]
+	var previous interactionSourceRefResponse
+	for index, ref := range refs {
+		if index > 0 && previous == ref {
+			continue
+		}
+		out = append(out, ref)
+		previous = ref
+	}
+	return out
+}
+
+func earliestSourcePosition(doc *source.SourceDocument, interaction interactionResponse) (int, bool) {
+	if doc == nil {
+		return 0, false
+	}
+	found := false
+	earliest := 0
+	for _, ref := range interaction.SourceRefs {
+		candidate := source.SourceRef{SourceID: doc.ID, StartUnitID: ref.StartUnitID, EndUnitID: ref.EndUnitID}
+		if err := source.ValidateRef(doc, candidate); err != nil {
+			continue
+		}
+		index, ok := source.UnitIndex(doc, candidate.StartUnitID)
+		if !ok || (found && index >= earliest) {
+			continue
+		}
+		earliest = index
+		found = true
+	}
+	return earliest, found
+}
+
+func unitSortValue(value int) int {
+	if value <= 0 {
+		return int(^uint(0) >> 1)
+	}
+	return value
+}
+
+func canonicalInteractionList(response extractionResponse, sourceID string) dnd.NPCInteractionList {
+	interactions := make([]dnd.NPCInteraction, len(response.Interactions))
+	for index, interaction := range response.Interactions {
+		interactions[index] = dnd.NPCInteraction{
+			Name:       interaction.Name,
+			Kind:       dnd.NPCInteractionKind(interaction.Kind),
+			SourceRefs: canonicalSourceRefs(interaction.SourceRefs, sourceID),
+		}
+	}
+	if response.Interactions == nil {
+		interactions = nil
+	}
+	return dnd.NPCInteractionList{Interactions: interactions}
+}
+
+func canonicalSourceRefs(refs []interactionSourceRefResponse, sourceID string) []source.SourceRef {
+	if refs == nil {
+		return nil
+	}
+	out := make([]source.SourceRef, len(refs))
+	for index, ref := range refs {
+		out[index] = source.SourceRef{SourceID: sourceID, StartUnitID: ref.StartUnitID, EndUnitID: ref.EndUnitID}
+	}
+	return out
+}

internal/modules/dnd/extract/npcinteractions/extractor.go

@@ -0,0 +1,225 @@
+package npcinteractions
+
+import (
+	"context"
+	"fmt"
+	"sort"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	npcregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/registry"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
+)
+
+const (
+	Key           = "dnd/npc-interactions"
+	mappingPolicy = "dnd.npc_interactions.extract_mapping.v1"
+)
+
+const (
+	NPCRegistryReferenceSlot = npcregistry.ReferenceSlot
+	NPCRegistryMaxBytes      = npcregistry.MaxBytes
+)
+
+var requiredCapabilities = []string{
+	"chunks",
+	"source.transcript",
+}
+
+var providedCapabilities = []string{
+	"dnd.npc_interactions",
+}
+
+var referenceSlotDescriptions = shared.ReferenceSlotDescriptions{
+	Glossary: "Optional campaign glossary reference material used only for interaction disambiguation.",
+	Party:    "Optional party roster reference material used only for interaction disambiguation.",
+	Players:  "Optional player list reference material used only for interaction disambiguation.",
+	Roster:   "Deprecated alias for party roster reference material used only for interaction disambiguation.",
+}
+
+func referenceSlots() []contracts.ReferenceSlot {
+	slots := shared.ReferenceSlots(referenceSlotDescriptions)
+	slots = append(slots, contracts.ReferenceSlot{
+		Name:                  NPCRegistryReferenceSlot,
+		Description:           "Required normalized NPC registry used only for interaction identity grounding, never as interaction evidence.",
+		Required:              true,
+		AcceptedMediaTypes:    []string{"application/json"},
+		AcceptedArtifactKinds: []contracts.ArtifactKind{dnd.NPCListKind},
+		MaxBytes:              NPCRegistryMaxBytes,
+	})
+	sort.Slice(slots, func(i, j int) bool { return slots[i].Name < slots[j].Name })
+	return slots
+}
+
+var _ contracts.Extractor[dnd.NPCInteractionList] = (*Extractor)(nil)
+var _ contracts.ManifestMetadataProvider = (*Extractor)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Extractor)(nil)
+
+type Options struct{}
+
+type Extractor struct {
+	llm               contracts.StructuredLLMClient
+	npcResolver       *npcregistry.Resolver
+	promptSHA         string
+	responseSchemaSHA string
+}
+
+func New(llmClient contracts.StructuredLLMClient, _ Options, references ...contracts.ReferenceSet) (*Extractor, error) {
+	if llmClient == nil {
+		return nil, extractorErrorf("LLM client must not be nil")
+	}
+	if len(references) > 1 {
+		return nil, extractorErrorf("at most one reference set may be supplied")
+	}
+	var referenceSet contracts.ReferenceSet
+	if len(references) == 1 {
+		referenceSet = references[0]
+	}
+	npcResolver, err := npcregistry.NewResolver(referenceSet)
+	if err != nil {
+		return nil, extractorErrorf("prepare NPC registry prompt input: %w", err)
+	}
+	promptSHA, err := scriptoriumPromptMetadata()
+	if err != nil {
+		return nil, extractorErrorf("load prompt metadata: %w", err)
+	}
+	responseSchema, err := loadResponseSchema()
+	if err != nil {
+		return nil, extractorErrorf("load response schema: %w", err)
+	}
+	return &Extractor{
+		llm:               llmClient,
+		npcResolver:       npcResolver,
+		promptSHA:         promptSHA,
+		responseSchemaSHA: responseSchema.SHA256,
+	}, nil
+}
+
+func (e *Extractor) Key() string { return Key }
+
+func (e *Extractor) ReferenceSlots() []contracts.ReferenceSlot { return referenceSlots() }
+
+func (e *Extractor) ManifestMetadata() map[string]any {
+	if e == nil {
+		return nil
+	}
+	metadata := map[string]any{
+		"prompt_id":               PromptID,
+		"prompt_version":          SchemaVersion,
+		"prompt_sha256":           e.promptSHA,
+		"mapping_policy":          mappingPolicy,
+		"response_schema_key":     string(ResponseSchemaKey),
+		"response_schema_id":      ResponseSchemaID,
+		"response_schema_name":    ResponseSchemaName,
+		"response_schema_version": SchemaVersion,
+		"response_schema_sha256":  e.responseSchemaSHA,
+	}
+	seeded := e.npcResolver.Seeded()
+	if seeded.Bound() {
+		metadata["npc_registry_digest"] = seeded.Digest()
+		metadata["npc_count"] = seeded.Count()
+	}
+	return metadata
+}
+
+func (e *Extractor) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	if e == nil {
+		return nil
+	}
+	seeded := e.npcResolver.Seeded()
+	return []pipeline.CheckpointFingerprint{
+		{Name: "prompt", Value: e.promptSHA},
+		{Name: "response_schema", Value: e.responseSchemaSHA},
+		{Name: "mapping_policy", Value: mappingPolicy},
+		{Name: "npc_registry", Value: seeded.ProjectionDigest()},
+	}
+}
+
+func (e *Extractor) Extract(ctx context.Context, req contracts.TypedExtractionRequest) (contracts.TypedExtractionResult[dnd.NPCInteractionList], error) {
+	if e == nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("extractor must not be nil")
+	}
+	if e.llm == nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("LLM client must not be nil")
+	}
+	if ctx == nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("context must not be nil")
+	}
+	if err := ctx.Err(); err != nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("context error before extraction: %w", err)
+	}
+	if req.Source == nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("source must not be nil")
+	}
+	if req.Chunk == nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("chunk must not be nil")
+	}
+	if len(req.Chunk.Units) == 0 {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("chunk %q units must not be empty", req.Chunk.ID)
+	}
+	sourceInput, err := shared.ChunkPromptMaterial(req)
+	if err != nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("%w", err)
+	}
+	npcRegistry, err := e.npcResolver.Resolve(req.References)
+	if err != nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("resolve NPC registry: %w", err)
+	}
+	if !npcRegistry.Bound() {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("NPC registry reference is required")
+	}
+
+	var response extractionResponse
+	inputs := shared.PromptInputs(sourceInput, req.References)
+	inputs[NPCRegistryReferenceSlot] = npcRegistry.PromptInput()
+	if _, err := e.llm.CompleteStructured(ctx, contracts.StructuredCompletionRequest{
+		StageName:     Key,
+		PromptID:      PromptID,
+		PromptVersion: SchemaVersion,
+		ProfileID:     req.LLMProfile,
+		SessionID:     req.SessionID,
+		Inputs:        inputs,
+	}, &response); err != nil {
+		return contracts.TypedExtractionResult[dnd.NPCInteractionList]{}, extractorErrorf("complete structured output: %w", err)
+	}
+	canonicalizeResponse(&response, req.Source)
+	return contracts.TypedExtractionResult[dnd.NPCInteractionList]{Value: canonicalInteractionList(response, req.Source.ID)}, nil
+}
+
+func ModuleSpec() pipeline.ModuleSpec {
+	return pipeline.ModuleSpec{
+		Key:            Key,
+		Stage:          pipeline.StageExtract,
+		Requires:       append([]string(nil), requiredCapabilities...),
+		Provides:       append([]string(nil), providedCapabilities...),
+		ArtifactKind:   dnd.NPCInteractionListKind,
+		ReferenceSlots: referenceSlots(),
+	}
+}
+
+func Register(registry *pipeline.ExtractorRegistry) error {
+	return pipeline.RegisterExtractorBuilder(registry, ModuleSpec(), validateOptions, func(request pipeline.BuildRequest) (contracts.Extractor[dnd.NPCInteractionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(request.Dependencies.LLM, options, request.References)
+	})
+}
+
+func validateOptions(options map[string]any) error {
+	_, err := DecodeOptions(options)
+	return err
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, extractorErrorf("%w", err)
+	}
+	return Options{}, nil
+}
+
+func extractorErrorf(format string, args ...any) error {
+	return fmt.Errorf("dnd NPC interactions extractor: "+format, args...)
+}

internal/modules/dnd/extract/npcinteractions/extractor_test.go

@@ -0,0 +1,370 @@
+package npcinteractions
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	npccodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcs"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
+)
+
+func TestExtractMapsEveryKindAndOrdersBySourcePosition(t *testing.T) {
+	client := &fakeInteractionsLLMClient{response: extractionResponse{Interactions: []interactionResponse{
+		{Name: "Other", Kind: "other", SourceRefs: interactionRefs(30, 30)},
+		{Name: "Opponent", Kind: "combat_opponent", SourceRefs: interactionRefs(20, 20)},
+		{Name: "Ally", Kind: "combat_ally", SourceRefs: interactionRefs(5, 5)},
+		{Name: "Speaker", Kind: "dialogue", SourceRefs: append(interactionRefs(2, 2), interactionRefs(2, 2)...)},
+		{Name: "Present", Kind: "noncombat_presence", SourceRefs: interactionRefs(7, 7)},
+		{Name: "Mentioned", Kind: "mentioned", SourceRefs: interactionRefs(10, 10)},
+		{Name: "Invalid", Kind: "unsupported", SourceRefs: interactionRefs(0, 0)},
+	}}}
+	references := requiredRegistryReferences(t, "Mentioned", "Speaker", "Present", "Ally", "Opponent", "Other", "Invalid")
+	req := extractionRequest()
+	req.References = references
+
+	result, err := newExtractor(t, client, references).Extract(context.Background(), req)
+	if err != nil {
+		t.Fatalf("Extract() error = %v", err)
+	}
+	if got := interactionNames(result.Value); !reflect.DeepEqual(got, []string{"Mentioned", "Speaker", "Present", "Ally", "Opponent", "Other", "Invalid"}) {
+		t.Fatalf("interaction order = %#v", got)
+	}
+	if got := interactionKinds(result.Value); !reflect.DeepEqual(got, []dnd.NPCInteractionKind{
+		dnd.NPCInteractionKindMentioned,
+		dnd.NPCInteractionKindDialogue,
+		dnd.NPCInteractionKindNoncombatPresence,
+		dnd.NPCInteractionKindCombatAlly,
+		dnd.NPCInteractionKindCombatOpponent,
+		dnd.NPCInteractionKindOther,
+		"unsupported",
+	}) {
+		t.Fatalf("interaction kinds = %#v", got)
+	}
+	if refs := result.Value.Interactions[1].SourceRefs; !reflect.DeepEqual(refs, []source.SourceRef{{SourceID: "session-alpha", StartUnitID: 2, EndUnitID: 2}}) {
+		t.Fatalf("canonical source refs = %#v", refs)
+	}
+	if invalid := result.Value.Interactions[6]; invalid.Name != "Invalid" || invalid.Kind != "unsupported" || !reflect.DeepEqual(invalid.SourceRefs, []source.SourceRef{{SourceID: "session-alpha"}}) {
+		t.Fatalf("invalid candidate = %#v, want preserved values with current source identity", invalid)
+	}
+	if len(client.requests) != 1 {
+		t.Fatalf("LLM calls = %d, want 1", len(client.requests))
+	}
+}
+
+func TestNewRequiresLLMAndRejectsAmbiguousReferenceSets(t *testing.T) {
+	if _, err := New(nil, Options{}); err == nil || !strings.Contains(err.Error(), "LLM client") {
+		t.Fatalf("New(nil) error = %v", err)
+	}
+	if _, err := New(&fakeInteractionsLLMClient{}, Options{}, contracts.ReferenceSet{}, contracts.ReferenceSet{}); err == nil || !strings.Contains(err.Error(), "at most one reference set") {
+		t.Fatalf("New() error = %v", err)
+	}
+}
+
+func TestExtractUsesNamesOnlyRegistryAndCurrentTranscriptEvidence(t *testing.T) {
+	client := &fakeInteractionsLLMClient{response: extractionResponse{Interactions: []interactionResponse{{
+		Name: "Mira Thorn", Kind: "dialogue", SourceRefs: interactionRefs(10, 10),
+	}}}}
+	references := requiredRegistryReferences(t, "Mira Thorn", "Hooded Guard")
+	req := extractionRequest()
+	req.References = references
+	if _, err := newExtractor(t, client, references).Extract(context.Background(), req); err != nil {
+		t.Fatalf("Extract() error = %v", err)
+	}
+	request := client.requests[0]
+	registry := request.Inputs[NPCRegistryReferenceSlot]
+	if registry.Name != NPCRegistryReferenceSlot || registry.MediaType != "application/json" || string(registry.Content) != `{"npcs":[{"name":"Mira Thorn"},{"name":"Hooded Guard"}]}` {
+		t.Fatalf("registry prompt input = %#v, want exact names-only projection", registry)
+	}
+	for _, forbidden := range []string{"npc:sha256:", "other-session", "start_unit_id"} {
+		if strings.Contains(string(registry.Content), forbidden) {
+			t.Fatalf("registry prompt input leaked %q: %s", forbidden, registry.Content)
+		}
+	}
+	if strings.Contains(string(request.Inputs["transcript"].Content), "other-session") {
+		t.Fatal("transcript input contains registry evidence")
+	}
+	metadata, err := json.Marshal(newExtractor(t, &fakeInteractionsLLMClient{}, references).ManifestMetadata())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if strings.Contains(string(metadata), "Mira Thorn") || strings.Contains(string(metadata), "other-session") {
+		t.Fatalf("manifest metadata leaked registry content: %s", metadata)
+	}
+}
+
+func TestExtractRequiresBoundRegistryBeforeLLMCall(t *testing.T) {
+	client := &fakeInteractionsLLMClient{}
+	if _, err := newExtractor(t, client).Extract(context.Background(), extractionRequest()); err == nil || !strings.Contains(err.Error(), "NPC registry reference is required") {
+		t.Fatalf("Extract() error = %v, want required registry failure", err)
+	}
+	if len(client.requests) != 0 {
+		t.Fatalf("LLM calls = %d, want no call", len(client.requests))
+	}
+
+	malformed := contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		NPCRegistryReferenceSlot: {
+			Slot:  contracts.ReferenceSlot{Name: NPCRegistryReferenceSlot},
+			Items: []contracts.ReferenceItem{{SlotName: NPCRegistryReferenceSlot, MediaType: "application/json", Content: []byte(`{"secret":"private source"}`)}},
+		},
+	}}
+	if _, err := New(client, Options{}, malformed); err == nil || !strings.Contains(err.Error(), "prepare NPC registry") || strings.Contains(err.Error(), "private source") {
+		t.Fatalf("New() error = %v, want content-safe malformed registry error", err)
+	}
+	if len(client.requests) != 0 {
+		t.Fatalf("LLM calls = %d, want no call", len(client.requests))
+	}
+}
+
+func TestExtractResolvesGeneratedRegistryAtOperationTime(t *testing.T) {
+	client := &fakeInteractionsLLMClient{response: extractionResponse{Interactions: []interactionResponse{}}}
+	references := requiredRegistryReferences(t, "Mira Thorn")
+	req := extractionRequest()
+	req.References = references
+	extractor := newExtractor(t, client)
+	if _, err := extractor.Extract(context.Background(), req); err != nil {
+		t.Fatalf("Extract() error = %v", err)
+	}
+	if input := client.requests[0].Inputs[NPCRegistryReferenceSlot]; string(input.Content) != `{"npcs":[{"name":"Mira Thorn"}]}` || input.OriginURI != "" {
+		t.Fatalf("generated registry prompt input = %#v", input)
+	}
+	metadata := extractor.ManifestMetadata()
+	if _, ok := metadata["npc_registry_digest"]; ok {
+		t.Fatalf("operation-time registry leaked into static metadata: %#v", metadata)
+	}
+}
+
+func TestExtractAcceptsEmptyBoundRegistryAndEmptyResponse(t *testing.T) {
+	content, err := npccodec.New().Encode(dnd.NPCList{NPCs: []dnd.NPC{}})
+	if err != nil {
+		t.Fatal(err)
+	}
+	references := contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		NPCRegistryReferenceSlot: {
+			Slot:  contracts.ReferenceSlot{Name: NPCRegistryReferenceSlot},
+			Items: []contracts.ReferenceItem{{SlotName: NPCRegistryReferenceSlot, MediaType: npccodec.MediaType, Content: content}},
+		},
+	}}
+	client := &fakeInteractionsLLMClient{response: extractionResponse{Interactions: []interactionResponse{}}}
+	req := extractionRequest()
+	req.References = references
+	result, err := newExtractor(t, client, references).Extract(context.Background(), req)
+	if err != nil || result.Value.Interactions == nil || len(result.Value.Interactions) != 0 {
+		t.Fatalf("Extract() = %#v, %v; want present empty interactions", result, err)
+	}
+}
+
+func TestExtractRejectsInvalidRequestsAndProviderFailures(t *testing.T) {
+	references := requiredRegistryReferences(t, "Mira Thorn")
+	valid := extractionRequest()
+	valid.References = references
+	canceled, cancel := context.WithCancel(context.Background())
+	cancel()
+	extractor := newExtractor(t, &fakeInteractionsLLMClient{}, references)
+	for _, test := range []struct {
+		name string
+		ctx  context.Context
+		req  contracts.TypedExtractionRequest
+		want string
+	}{
+		{"nil context", nil, valid, "context"},
+		{"canceled context", canceled, valid, "context"},
+		{"nil source", context.Background(), contracts.TypedExtractionRequest{Chunk: valid.Chunk, References: references}, "source"},
+		{"nil chunk", context.Background(), contracts.TypedExtractionRequest{Source: valid.Source, References: references}, "chunk"},
+		{"empty chunk", context.Background(), emptyChunkRequest(valid), "units"},
+		{"source input mismatch", context.Background(), mismatchedSourceInputRequest(valid), "must match chunk"},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := extractor.Extract(test.ctx, test.req); err == nil || !strings.Contains(err.Error(), test.want) {
+				t.Fatalf("Extract() error = %v, want %q", err, test.want)
+			}
+		})
+	}
+	if _, err := newExtractor(t, &fakeInteractionsLLMClient{err: errors.New("provider unavailable")}, references).Extract(context.Background(), valid); err == nil || !strings.Contains(err.Error(), "provider unavailable") {
+		t.Fatalf("provider error = %v", err)
+	}
+}
+
+func TestModuleSpecRegistrationMetadataAndFingerprints(t *testing.T) {
+	got := ModuleSpec()
+	if got.Key != Key || got.Stage != pipeline.StageExtract || got.ArtifactKind != dnd.NPCInteractionListKind || !reflect.DeepEqual(got.Requires, []string{"chunks", "source.transcript"}) || !reflect.DeepEqual(got.Provides, []string{"dnd.npc_interactions"}) {
+		t.Fatalf("ModuleSpec() = %#v", got)
+	}
+	var registrySlot contracts.ReferenceSlot
+	for _, slot := range got.ReferenceSlots {
+		if slot.Name == NPCRegistryReferenceSlot {
+			registrySlot = slot
+		}
+	}
+	if !registrySlot.Required || !reflect.DeepEqual(registrySlot.AcceptedArtifactKinds, []contracts.ArtifactKind{dnd.NPCListKind}) || registrySlot.MaxBytes != NPCRegistryMaxBytes {
+		t.Fatalf("NPC registry slot = %#v", registrySlot)
+	}
+	got.ReferenceSlots[0].AcceptedMediaTypes[0] = "changed"
+	if ModuleSpec().ReferenceSlots[0].AcceptedMediaTypes[0] == "changed" {
+		t.Fatal("ModuleSpec() returned mutable reference slots")
+	}
+	extractorSlots := newExtractor(t, &fakeInteractionsLLMClient{}).ReferenceSlots()
+	extractorSlots[0].AcceptedMediaTypes[0] = "changed"
+	if newExtractor(t, &fakeInteractionsLLMClient{}).ReferenceSlots()[0].AcceptedMediaTypes[0] == "changed" {
+		t.Fatal("ReferenceSlots() returned mutable reference slots")
+	}
+	registry := pipeline.NewExtractorRegistry()
+	if err := Register(registry); err != nil {
+		t.Fatal(err)
+	}
+	if _, ok := registry.Spec(Key); !ok {
+		t.Fatalf("registry missing %q", Key)
+	}
+	if _, err := DecodeOptions(map[string]any{"unexpected": true}); err == nil {
+		t.Fatal("DecodeOptions() accepted unknown option")
+	}
+
+	references := requiredRegistryReferences(t, "Mira Thorn")
+	extractor := newExtractor(t, &fakeInteractionsLLMClient{}, references)
+	metadata := extractor.ManifestMetadata()
+	for key, want := range map[string]string{
+		"prompt_id": PromptID, "prompt_version": SchemaVersion, "mapping_policy": mappingPolicy,
+		"response_schema_key": string(ResponseSchemaKey), "response_schema_id": ResponseSchemaID,
+		"response_schema_name": ResponseSchemaName, "response_schema_version": SchemaVersion,
+	} {
+		if metadata[key] != want {
+			t.Fatalf("metadata[%q] = %#v, want %q", key, metadata[key], want)
+		}
+	}
+	for _, key := range []string{"prompt_sha256", "response_schema_sha256", "npc_registry_digest"} {
+		if value, ok := metadata[key].(string); !ok || !strings.HasPrefix(value, "sha256:") {
+			t.Fatalf("metadata[%q] = %#v, want digest", key, metadata[key])
+		}
+	}
+	wantFingerprints := map[string]struct{}{"prompt": {}, "response_schema": {}, "mapping_policy": {}, "npc_registry": {}}
+	for _, fingerprint := range extractor.CheckpointFingerprints() {
+		delete(wantFingerprints, fingerprint.Name)
+	}
+	if len(wantFingerprints) != 0 {
+		t.Fatalf("missing fingerprints = %#v", wantFingerprints)
+	}
+}
+
+func extractionRequest() contracts.TypedExtractionRequest {
+	doc := sourceDocument()
+	chunk := &source.Chunk{
+		ID: "session-alpha:chunk:0", SourceID: doc.ID, Index: 0,
+		Ref:     source.SourceRef{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 30},
+		Content: []byte(`{"units":[10,2,7,5,20,30]}`), MediaType: "application/json",
+		Units: append([]source.SourceUnit(nil), doc.Units...),
+	}
+	return contracts.TypedExtractionRequest{
+		Source: doc, Chunk: chunk,
+		SourceInput: contracts.NewLLMInputMaterial("source", chunk.MediaType, chunk.Content, "sha256:chunk", "file:///session-alpha.json"),
+		SessionID:   "session-123", LLMProfile: "profile-npc-interactions",
+	}
+}
+
+func sourceDocument() *source.SourceDocument {
+	return &source.SourceDocument{
+		ID: "session-alpha", Kind: "transcript", Format: "application/vnd.seriatim.minimal+json", Digest: "sha256:source",
+		Units: []source.SourceUnit{
+			{ID: 10, Kind: "transcript_segment", Text: "Mira Thorn is mentioned."},
+			{ID: 2, Kind: "transcript_segment", Text: "Mira Thorn speaks to the party."},
+			{ID: 7, Kind: "transcript_segment", Text: "Mira Thorn watches nearby."},
+			{ID: 5, Kind: "transcript_segment", Text: "Mira Thorn joins the party in combat."},
+			{ID: 20, Kind: "transcript_segment", Text: "Mira Thorn attacks the party."},
+			{ID: 30, Kind: "transcript_segment", Text: "Mira Thorn performs a ritual."},
+		},
+	}
+}
+
+func requiredRegistryReferences(t *testing.T, names ...string) contracts.ReferenceSet {
+	t.Helper()
+	npcs := make([]dnd.NPC, len(names))
+	for index, name := range names {
+		npcs[index] = dnd.NPC{
+			ID: identity.DeriveID(name), Name: name,
+			SourceRefs: []source.SourceRef{{SourceID: "other-session", StartUnitID: index + 1, EndUnitID: index + 1}},
+		}
+	}
+	content, err := npccodec.New().Encode(dnd.NPCList{NPCs: npcs})
+	if err != nil {
+		t.Fatal(err)
+	}
+	return contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		NPCRegistryReferenceSlot: {
+			Slot:  contracts.ReferenceSlot{Name: NPCRegistryReferenceSlot},
+			Items: []contracts.ReferenceItem{{SlotName: NPCRegistryReferenceSlot, MediaType: npccodec.MediaType, Content: content, Origin: contracts.ReferenceOrigin{Type: "generated"}}},
+		},
+	}}
+}
+
+func interactionRefs(start, end int) []interactionSourceRefResponse {
+	return []interactionSourceRefResponse{{StartUnitID: start, EndUnitID: end}}
+}
+
+func interactionNames(value dnd.NPCInteractionList) []string {
+	names := make([]string, len(value.Interactions))
+	for index, interaction := range value.Interactions {
+		names[index] = interaction.Name
+	}
+	return names
+}
+
+func interactionKinds(value dnd.NPCInteractionList) []dnd.NPCInteractionKind {
+	kinds := make([]dnd.NPCInteractionKind, len(value.Interactions))
+	for index, interaction := range value.Interactions {
+		kinds[index] = interaction.Kind
+	}
+	return kinds
+}
+
+func emptyChunkRequest(req contracts.TypedExtractionRequest) contracts.TypedExtractionRequest {
+	req.Chunk = &source.Chunk{ID: req.Chunk.ID, SourceID: req.Chunk.SourceID, Index: req.Chunk.Index}
+	return req
+}
+
+func mismatchedSourceInputRequest(req contracts.TypedExtractionRequest) contracts.TypedExtractionRequest {
+	req.SourceInput = contracts.NewLLMInputMaterial("source", "application/json", []byte(`{"different":true}`), "sha256:other", "")
+	return req
+}
+
+func newExtractor(t *testing.T, client contracts.StructuredLLMClient, references ...contracts.ReferenceSet) *Extractor {
+	t.Helper()
+	extractor, err := New(client, Options{}, references...)
+	if err != nil {
+		t.Fatalf("New() error = %v", err)
+	}
+	return extractor
+}
+
+type fakeInteractionsLLMClient struct {
+	response extractionResponse
+	err      error
+	requests []contracts.StructuredCompletionRequest
+}
+
+func (client *fakeInteractionsLLMClient) CompleteStructured(_ context.Context, req contracts.StructuredCompletionRequest, out any) (contracts.StructuredCompletionResponse, error) {
+	client.requests = append(client.requests, cloneStructuredCompletionRequest(req))
+	if client.err != nil {
+		return contracts.StructuredCompletionResponse{}, client.err
+	}
+	target, ok := out.(*extractionResponse)
+	if !ok {
+		return contracts.StructuredCompletionResponse{}, errors.New("unexpected output target")
+	}
+	*target = client.response
+	content, err := json.Marshal(client.response)
+	if err != nil {
+		return contracts.StructuredCompletionResponse{}, err
+	}
+	return contracts.StructuredCompletionResponse{Content: content}, nil
+}
+
+func cloneStructuredCompletionRequest(req contracts.StructuredCompletionRequest) contracts.StructuredCompletionRequest {
+	req.Inputs = req.Inputs.Clone()
+	return req
+}

internal/modules/dnd/extract/npcinteractions/model.go

@@ -0,0 +1,16 @@
+package npcinteractions
+
+type extractionResponse struct {
+	Interactions []interactionResponse `json:"interactions"`
+}
+
+type interactionResponse struct {
+	Name       string                         `json:"name"`
+	Kind       string                         `json:"kind"`
+	SourceRefs []interactionSourceRefResponse `json:"source_refs"`
+}
+
+type interactionSourceRefResponse struct {
+	StartUnitID int `json:"start_unit_id"`
+	EndUnitID   int `json:"end_unit_id"`
+}

internal/modules/dnd/extract/npcinteractions/model_test.go

@@ -0,0 +1,18 @@
+package npcinteractions
+
+import (
+	"encoding/json"
+	"testing"
+)
+
+func TestExtractionResponsePreservesValidatorOwnedSemantics(t *testing.T) {
+	content := []byte(`{"interactions":[{"name":"","kind":"unsupported","source_refs":[{"start_unit_id":0,"end_unit_id":-1}]}]}`)
+	var response extractionResponse
+	if err := json.Unmarshal(content, &response); err != nil {
+		t.Fatalf("json.Unmarshal() error = %v", err)
+	}
+	interaction := response.Interactions[0]
+	if interaction.Name != "" || interaction.Kind != "unsupported" || interaction.SourceRefs[0] != (interactionSourceRefResponse{StartUnitID: 0, EndUnitID: -1}) {
+		t.Fatalf("decoded response = %#v", interaction)
+	}
+}

internal/modules/dnd/extract/npcinteractions/schema.go

@@ -0,0 +1,21 @@
+package npcinteractions
+
+import "gitea.maximumdirect.net/eric/notarius/internal/framework/llm"
+
+const (
+	PromptID           = "dnd.npc_interactions"
+	ResponseSchemaKey  = llm.ResponseSchemaKey("dnd_npc_interactions_llm")
+	ResponseSchemaID   = "notarius.dnd.npc_interactions.llm"
+	ResponseSchemaName = "notarius_dnd_npc_interactions_llm_v1"
+	SchemaVersion      = "v1"
+)
+
+func loadResponseSchema() (llm.ResponseSchema, error) {
+	return llm.LoadResponseSchema(embeddedAssets, llm.ResponseSchemaDefinition{
+		Key:       ResponseSchemaKey,
+		ID:        ResponseSchemaID,
+		Version:   SchemaVersion,
+		Name:      ResponseSchemaName,
+		AssetPath: "assets/schemas/dnd_npc_interactions_llm.v1.json",
+	})
+}

internal/modules/dnd/extract/npcinteractions/schema_test.go

@@ -0,0 +1,106 @@
+package npcinteractions
+
+import (
+	"bytes"
+	"encoding/json"
+	"strings"
+	"testing"
+
+	"github.com/santhosh-tekuri/jsonschema/v6"
+)
+
+func TestResponseSchemaOwnsOnlyPrivateStructuralContract(t *testing.T) {
+	schema, err := loadResponseSchema()
+	if err != nil {
+		t.Fatal(err)
+	}
+	if schema.Key != ResponseSchemaKey || schema.ID != ResponseSchemaID || schema.Name != ResponseSchemaName || schema.Version != SchemaVersion || !strings.HasPrefix(schema.SHA256, "sha256:") || !json.Valid(schema.JSONSchema) {
+		t.Fatalf("schema = %#v", schema)
+	}
+	valid := validInteractionResponse()
+	content, err := json.Marshal(valid)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := validateJSONSchema(content, schema.JSONSchema); err != nil {
+		t.Fatalf("valid response rejected: %v", err)
+	}
+
+	semanticCandidate := validInteractionResponse()
+	interaction := semanticCandidate["interactions"].([]any)[0].(map[string]any)
+	interaction["name"] = ""
+	interaction["kind"] = "unsupported"
+	ref := interaction["source_refs"].([]any)[0].(map[string]any)
+	ref["start_unit_id"] = 0
+	ref["end_unit_id"] = -1
+	content, err = json.Marshal(semanticCandidate)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := validateJSONSchema(content, schema.JSONSchema); err != nil {
+		t.Fatalf("schema rejected validator-owned semantics: %v", err)
+	}
+
+	for _, mutate := range []func(map[string]any){
+		func(record map[string]any) { delete(record, "name") },
+		func(record map[string]any) { record["kind"] = 1 },
+		func(record map[string]any) { record["unexpected"] = true },
+		func(record map[string]any) {
+			record["source_refs"].([]any)[0].(map[string]any)["source_id"] = "assigned later"
+		},
+	} {
+		candidate := validInteractionResponse()
+		mutate(candidate["interactions"].([]any)[0].(map[string]any))
+		content, err := json.Marshal(candidate)
+		if err != nil {
+			t.Fatal(err)
+		}
+		if err := validateJSONSchema(content, schema.JSONSchema); err == nil {
+			t.Fatal("schema accepted structurally invalid response")
+		}
+	}
+}
+
+func TestResponseSchemaIsDefensiveAndContentSafe(t *testing.T) {
+	first, err := loadResponseSchema()
+	if err != nil {
+		t.Fatal(err)
+	}
+	first.JSONSchema[0] = '['
+	second, err := loadResponseSchema()
+	if err != nil || !json.Valid(second.JSONSchema) || bytes.Equal(first.JSONSchema, second.JSONSchema) {
+		t.Fatalf("second schema = %s, %v", second.JSONSchema, err)
+	}
+	if diagnostics := second.DiagnosticsMap(); diagnostics["key"] != ResponseSchemaKey || diagnostics["id"] != ResponseSchemaID {
+		t.Fatalf("diagnostics = %#v", diagnostics)
+	} else if _, ok := diagnostics["json_schema"]; ok {
+		t.Fatalf("diagnostics leak schema content: %#v", diagnostics)
+	}
+}
+
+func validInteractionResponse() map[string]any {
+	return map[string]any{"interactions": []any{map[string]any{
+		"name": "Mira Thorn", "kind": "dialogue",
+		"source_refs": []any{map[string]any{"start_unit_id": 1, "end_unit_id": 2}},
+	}}}
+}
+
+func validateJSONSchema(instanceContent, schemaContent []byte) error {
+	instance, err := jsonschema.UnmarshalJSON(bytes.NewReader(instanceContent))
+	if err != nil {
+		return err
+	}
+	schemaDocument, err := jsonschema.UnmarshalJSON(bytes.NewReader(schemaContent))
+	if err != nil {
+		return err
+	}
+	compiler := jsonschema.NewCompiler()
+	if err := compiler.AddResource("schema.json", schemaDocument); err != nil {
+		return err
+	}
+	schema, err := compiler.Compile("schema.json")
+	if err != nil {
+		return err
+	}
+	return schema.Validate(instance)
+}

internal/modules/dnd/extract/npcinteractions/scriptorium_assets.go

@@ -0,0 +1,53 @@
+package npcinteractions
+
+import (
+	"fmt"
+	"sync"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/llm"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/promptfs"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
+)
+
+const scriptoriumPromptRoot = "assets/prompts"
+
+var promptAssetManifest = shared.PromptAssetManifest{
+	ModuleDir: "dnd.npc_interactions",
+	ModuleFiles: []promptfs.ModulePromptFile{
+		{Name: "dnd.npc_interactions.yaml", Path: "assets/prompts/dnd.npc_interactions.yaml"},
+		{Name: "task.md", Path: "assets/prompts/task.md"},
+		{Name: "instructions.md", Path: "assets/prompts/instructions.md"},
+	},
+	SharedFiles: []string{
+		"common-dnd-system.md",
+		"common-dnd-extraction-evidence.md",
+		"common-dnd-identity.md",
+		"common-dnd-transcript.md",
+		"common-dnd-references.md",
+		"common-dnd-npcs.md",
+	},
+}
+
+func RegisterPromptAssets(registry *llm.AssetRegistry) error {
+	promptFS, err := promptAssetManifest.PromptFS(embeddedAssets)
+	if err != nil {
+		return fmt.Errorf("prepare NPC-interaction prompt assets: %w", err)
+	}
+	if err := registry.RegisterPromptFS(promptFS, scriptoriumPromptRoot); err != nil {
+		return err
+	}
+	return registry.RegisterSchemaFS(embeddedAssets, "assets/schemas")
+}
+
+func scriptoriumPromptMetadata() (string, error) {
+	scriptoriumPromptHashOnce.Do(func() {
+		scriptoriumPromptHash, scriptoriumPromptHashErr = promptAssetManifest.Hash(embeddedAssets)
+	})
+	return scriptoriumPromptHash, scriptoriumPromptHashErr
+}
+
+var (
+	scriptoriumPromptHashOnce sync.Once
+	scriptoriumPromptHash     string
+	scriptoriumPromptHashErr  error
+)

internal/modules/dnd/extract/npcinteractions/scriptorium_assets_test.go

@@ -0,0 +1,94 @@
+package npcinteractions
+
+import (
+	"context"
+	"io/fs"
+	"strings"
+	"testing"
+	"time"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/llm"
+	"gitea.maximumdirect.net/eric/scriptorium"
+)
+
+func TestRegisterPromptAssetsAndPrepareInteractionPrompt(t *testing.T) {
+	registry := llm.NewAssetRegistry()
+	if err := RegisterPromptAssets(registry); err != nil {
+		t.Fatal(err)
+	}
+	schemaFS, err := registry.SchemaFS()
+	if err != nil {
+		t.Fatal(err)
+	}
+	if _, err := fs.ReadFile(schemaFS, "dnd_npc_interactions_llm.v1.json"); err != nil {
+		t.Fatalf("response schema asset: %v", err)
+	}
+	options, err := registry.ScriptoriumOptions()
+	if err != nil {
+		t.Fatal(err)
+	}
+	options = append(options, scriptorium.WithProfiles(scriptorium.OpenAICompatibleProfile(scriptorium.OpenAICompatibleProfileConfig{
+		ID: "npc-interactions-test-profile", Endpoint: "http://127.0.0.1:1/v1", Model: "npc-interactions-test-model",
+	})))
+	engine, err := scriptorium.NewEngine(scriptorium.Config{Timeout: time.Second}, options...)
+	if err != nil {
+		t.Fatal(err)
+	}
+	transcript := `{"units":[1]}`
+	prepared, err := engine.Prepare(context.Background(), scriptorium.RunRequest{
+		PromptID: PromptID, PromptVersion: SchemaVersion, ProfileID: "npc-interactions-test-profile",
+		Inputs: map[string]scriptorium.ArtifactRef{
+			"transcript": scriptorium.InlineWithURI("file:///session.json", transcript),
+			"players":    scriptorium.Inline("Dana: Mira"),
+			"party":      scriptorium.Inline("Mira: ranger"),
+			"glossary":   scriptorium.Inline("Greencloak: title"),
+			"npcs":       scriptorium.Inline(`{"npcs":[{"name":"Mira Thorn"}]}`),
+		},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if prepared.PromptID != PromptID || prepared.OutputContract.SchemaPath != "dnd_npc_interactions_llm.v1.json" {
+		t.Fatalf("prepared prompt = %#v", prepared)
+	}
+	for _, want := range []string{
+		"Mira Thorn", "mentioned", "combat_opponent", "Registry content is context, not event evidence", transcript,
+	} {
+		found := false
+		for _, message := range prepared.Messages {
+			if strings.Contains(message.Content, want) {
+				found = true
+				break
+			}
+		}
+		if !found {
+			t.Fatalf("prepared prompt did not include %q", want)
+		}
+	}
+	if last := prepared.Messages[len(prepared.Messages)-1]; !strings.Contains(last.Content, transcript) {
+		t.Fatalf("last prompt message = %q, want transcript", last.Content)
+	}
+}
+
+func TestPromptMetadataDoesNotExposeAssetContent(t *testing.T) {
+	hash, err := scriptoriumPromptMetadata()
+	if err != nil || !strings.HasPrefix(hash, "sha256:") {
+		t.Fatalf("scriptoriumPromptMetadata() = %q, %v", hash, err)
+	}
+	metadata := newExtractor(t, &fakeInteractionsLLMClient{}).ManifestMetadata()
+	for _, forbidden := range []string{"combat_opponent", "common-dnd-system", "source_refs", "dnd_npc_interactions_llm.v1.json"} {
+		if strings.Contains(strings.Join(mapValues(metadata), " "), forbidden) {
+			t.Fatalf("metadata leaked prompt or schema content %q: %#v", forbidden, metadata)
+		}
+	}
+}
+
+func mapValues(values map[string]any) []string {
+	out := make([]string, 0, len(values))
+	for _, value := range values {
+		if text, ok := value.(string); ok {
+			out = append(out, text)
+		}
+	}
+	return out
+}

internal/modules/dnd/extract/scenedescriptions/assets.go

@@ -0,0 +1,6 @@
+package scenedescriptions
+
+import "embed"
+
+//go:embed assets/schemas/dnd_scene_descriptions_llm.v1.json assets/prompts/*.yaml assets/prompts/*.md
+var embeddedAssets embed.FS

internal/modules/dnd/extract/scenedescriptions/assets/prompts/dnd.scene_descriptions.yaml

@@ -0,0 +1,40 @@
+id: dnd.scene_descriptions
+version: "v1"
+default_profile: gemini-2-flash
+inputs:
+  - name: transcript
+    required: true
+    content_type: application/json
+  - name: players
+    required: false
+    content_type: text/plain
+  - name: party
+    required: false
+    content_type: text/plain
+  - name: glossary
+    required: false
+    content_type: text/plain
+messages:
+  - role: system
+    content_file: ./sharedassets/common-dnd-system.md
+  - role: user
+    content_file: ./sharedassets/common-dnd-identity.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./sharedassets/common-dnd-references.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./task.md
+  - role: user
+    content_file: ./instructions.md
+    cache_control:
+      type: ephemeral
+  - role: user
+    content_file: ./sharedassets/common-dnd-transcript.md
+output:
+  format: json
+  validation_mode: json_schema
+  schema_path: dnd_scene_descriptions_llm.v1.json
+  repair_attempts: 0

internal/modules/dnd/extract/scenedescriptions/assets/prompts/instructions.md

@@ -0,0 +1,44 @@
+Choose exactly one kind:
+
+- combat: active combat materially organizes the scene, including
+  initiative-like exchanges or sustained hostile action. Planning a fight or
+  discussing a completed fight is not combat by itself.
+- narrative: current-session in-world play that is not principally active
+  combat, a prior-session recap, or sustained out-of-character session
+  discussion. This includes exploration, travel, dialogue, investigation,
+  in-character planning, and aftermath.
+- recap: the scene's organizing purpose is to recount events from a previous
+  session for the table. An in-world character recounting history during
+  current play remains narrative.
+- meta: the scene's organizing purpose is sustained out-of-character
+  discussion about the game or session rather than advancing current in-world
+  play.
+
+Narrative is the default for actual current-session gameplay that does not meet
+another definition. When the accepted chunk is mixed:
+
+1. use combat when active combat is a substantive central activity, even with
+   brief setup, rules clarification, or immediate aftermath;
+2. otherwise use recap when recounting a previous session is the chunk's
+   primary table purpose;
+3. otherwise use meta when sustained out-of-character session discussion is
+   primary and in-world progression is no more than incidental; and
+4. use narrative for all remaining current-session in-world play.
+
+Brief table talk, dice resolution, rules clarification, jokes, or
+administrative comments do not make a gameplay scene meta. A short recollection
+used to orient current action does not make a scene recap.
+
+The title must be a short, distinguishing phrase rather than a sentence,
+chapter number, or generic label such as "Scene." It may use names and places
+established by the transcript or disambiguated by campaign references, but it
+must not invent a proper noun.
+
+The summary must briefly state the main activity and material transition or
+outcome established within the accepted chunk. Do not add analysis, inferred
+motives, hidden state, future consequences, relationship claims, or facts from
+outside the chunk. Campaign references may disambiguate names but never add
+events or lore.
+
+Do not return identifiers, source identifiers, source ranges, unit identifiers,
+participants, confidence, or any fields besides kind, title, and summary.

internal/modules/dnd/extract/scenedescriptions/assets/prompts/task.md

@@ -0,0 +1,5 @@
+Describe exactly one accepted Dungeons & Dragons scene from the supplied
+transcript chunk. The complete chunk is the evidence boundary: do not split it
+into multiple scenes or use facts that are not supported by it.
+
+Return one kind, one concise title, and one concise summary.

internal/modules/dnd/extract/scenedescriptions/assets/schemas/dnd_scene_descriptions_llm.v1.json

@@ -0,0 +1,21 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "notarius.dnd.scene_descriptions.llm",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["kind", "title", "summary"],
+  "properties": {
+    "kind": {
+      "type": "string",
+      "enum": ["combat", "narrative", "recap", "meta"]
+    },
+    "title": {
+      "type": "string",
+      "minLength": 1
+    },
+    "summary": {
+      "type": "string",
+      "minLength": 1
+    }
+  }
+}

internal/modules/dnd/extract/scenedescriptions/extractor.go

@@ -0,0 +1,195 @@
+package scenedescriptions
+
+import (
+	"context"
+	"fmt"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
+)
+
+const Key = "dnd/scene-descriptions"
+
+var requiredCapabilities = []string{
+	"chunks",
+	"source.transcript",
+}
+
+var providedCapabilities = []string{
+	"dnd.scene_descriptions",
+}
+
+var referenceSlotDescriptions = shared.ReferenceSlotDescriptions{
+	Glossary: "Optional campaign glossary reference material used only to disambiguate scene descriptions.",
+	Party:    "Optional party roster reference material used only to disambiguate scene descriptions.",
+	Players:  "Optional player list reference material used only to disambiguate scene descriptions.",
+}
+
+func referenceSlots() []contracts.ReferenceSlot {
+	all := shared.ReferenceSlots(referenceSlotDescriptions)
+	slots := make([]contracts.ReferenceSlot, 0, 3)
+	for _, slot := range all {
+		if slot.Name != "roster" {
+			slots = append(slots, slot)
+		}
+	}
+	return slots
+}
+
+var _ contracts.Extractor[dnd.SceneDescriptionList] = (*Extractor)(nil)
+var _ contracts.ManifestMetadataProvider = (*Extractor)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Extractor)(nil)
+
+type Options struct{}
+
+type Extractor struct {
+	llm               contracts.StructuredLLMClient
+	promptSHA         string
+	responseSchemaSHA string
+}
+
+func New(llmClient contracts.StructuredLLMClient, _ Options, references ...contracts.ReferenceSet) (*Extractor, error) {
+	if llmClient == nil {
+		return nil, extractorErrorf("LLM client must not be nil")
+	}
+	if len(references) > 1 {
+		return nil, extractorErrorf("at most one reference set may be supplied")
+	}
+	promptSHA, err := scriptoriumPromptMetadata()
+	if err != nil {
+		return nil, extractorErrorf("load prompt metadata: %w", err)
+	}
+	responseSchema, err := loadResponseSchema()
+	if err != nil {
+		return nil, extractorErrorf("load response schema: %w", err)
+	}
+	return &Extractor{
+		llm:               llmClient,
+		promptSHA:         promptSHA,
+		responseSchemaSHA: responseSchema.SHA256,
+	}, nil
+}
+
+func (e *Extractor) Key() string { return Key }
+
+func (e *Extractor) ReferenceSlots() []contracts.ReferenceSlot { return referenceSlots() }
+
+func (e *Extractor) ManifestMetadata() map[string]any {
+	if e == nil {
+		return nil
+	}
+	return map[string]any{
+		"prompt_id":               PromptID,
+		"prompt_version":          SchemaVersion,
+		"prompt_sha256":           e.promptSHA,
+		"response_schema_key":     string(ResponseSchemaKey),
+		"response_schema_id":      ResponseSchemaID,
+		"response_schema_name":    ResponseSchemaName,
+		"response_schema_version": SchemaVersion,
+		"response_schema_sha256":  e.responseSchemaSHA,
+		"mapping_policy":          mappingPolicy,
+	}
+}
+
+func (e *Extractor) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	if e == nil {
+		return nil
+	}
+	return []pipeline.CheckpointFingerprint{
+		{Name: "prompt", Value: e.promptSHA},
+		{Name: "response_schema", Value: e.responseSchemaSHA},
+		{Name: "mapping_policy", Value: mappingPolicy},
+	}
+}
+
+func (e *Extractor) Extract(ctx context.Context, req contracts.TypedExtractionRequest) (contracts.TypedExtractionResult[dnd.SceneDescriptionList], error) {
+	if e == nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("extractor must not be nil")
+	}
+	if e.llm == nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("LLM client must not be nil")
+	}
+	if ctx == nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("context must not be nil")
+	}
+	if err := ctx.Err(); err != nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("context error before extraction: %w", err)
+	}
+	if req.Source == nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("source must not be nil")
+	}
+	if req.Chunk == nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("chunk must not be nil")
+	}
+	if len(req.Chunk.Units) == 0 {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("chunk %q units must not be empty", req.Chunk.ID)
+	}
+	sourceInput, err := shared.ChunkPromptMaterial(req)
+	if err != nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("%w", err)
+	}
+
+	var response extractionResponse
+	if _, err := e.llm.CompleteStructured(ctx, contracts.StructuredCompletionRequest{
+		StageName:     Key,
+		PromptID:      PromptID,
+		PromptVersion: SchemaVersion,
+		ProfileID:     req.LLMProfile,
+		SessionID:     req.SessionID,
+		Inputs:        shared.PromptInputs(sourceInput, req.References),
+	}, &response); err != nil {
+		return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{}, extractorErrorf("complete structured output: %w", err)
+	}
+	return contracts.TypedExtractionResult[dnd.SceneDescriptionList]{Value: mapResponse(response, req.Chunk)}, nil
+}
+
+func mapResponse(response extractionResponse, chunk *source.Chunk) dnd.SceneDescriptionList {
+	return dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{
+		ID:        chunk.ID,
+		SourceRef: chunk.Ref,
+		Kind:      response.Kind,
+		Title:     strings.TrimSpace(response.Title),
+		Summary:   strings.TrimSpace(response.Summary),
+	}}}
+}
+
+func ModuleSpec() pipeline.ModuleSpec {
+	return pipeline.ModuleSpec{
+		Key:            Key,
+		Stage:          pipeline.StageExtract,
+		Requires:       append([]string(nil), requiredCapabilities...),
+		Provides:       append([]string(nil), providedCapabilities...),
+		ArtifactKind:   dnd.SceneDescriptionListKind,
+		ReferenceSlots: referenceSlots(),
+	}
+}
+
+func Register(registry *pipeline.ExtractorRegistry) error {
+	return pipeline.RegisterExtractorBuilder(registry, ModuleSpec(), validateOptions, func(request pipeline.BuildRequest) (contracts.Extractor[dnd.SceneDescriptionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(request.Dependencies.LLM, options, request.References)
+	})
+}
+
+func validateOptions(options map[string]any) error {
+	_, err := DecodeOptions(options)
+	return err
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, extractorErrorf("%w", err)
+	}
+	return Options{}, nil
+}
+
+func extractorErrorf(format string, args ...any) error {
+	return fmt.Errorf("dnd scene descriptions extractor: "+format, args...)
+}

internal/modules/dnd/extract/scenedescriptions/extractor_test.go

@@ -0,0 +1,126 @@
+package scenedescriptions
+
+import (
+	"context"
+	"errors"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+func TestExtractMapsExactlyOneSceneToTheChunk(t *testing.T) {
+	client := &fakeSceneDescriptionsLLMClient{response: extractionResponse{
+		Kind: dnd.SceneKindCombat, Title: "  Fight at the Watchtower  ", Summary: "  Bandits attack the party.  ",
+	}}
+	request := extractionRequest()
+	result, err := newExtractor(t, client).Extract(context.Background(), request)
+	if err != nil {
+		t.Fatalf("Extract() error = %v, want nil", err)
+	}
+	want := dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{
+		ID: request.Chunk.ID, SourceRef: request.Chunk.Ref, Kind: dnd.SceneKindCombat,
+		Title: "Fight at the Watchtower", Summary: "Bandits attack the party.",
+	}}}
+	if !reflect.DeepEqual(result.Value, want) {
+		t.Fatalf("Value = %#v, want %#v", result.Value, want)
+	}
+	if len(client.requests) != 1 {
+		t.Fatalf("LLM calls = %d, want 1", len(client.requests))
+	}
+	completed := client.requests[0]
+	if completed.StageName != Key || completed.PromptID != PromptID || completed.PromptVersion != SchemaVersion || completed.ProfileID != "profile-scene-descriptions" || completed.SessionID != "session-123" {
+		t.Fatalf("LLM request identity = %#v", completed)
+	}
+	transcript := completed.Inputs["transcript"]
+	if transcript.Name != "transcript" || transcript.MediaType != "application/json" || transcript.Digest != "sha256:chunk" || transcript.OriginURI != "file:///session-alpha.json" || string(transcript.Content) != string(request.Chunk.Content) {
+		t.Fatalf("transcript input = %#v, want chunk-scoped material", transcript)
+	}
+}
+
+func TestExtractPassesOptionalReferencesAndUsesEmptyPlaceholders(t *testing.T) {
+	client := &fakeSceneDescriptionsLLMClient{response: extractionResponse{Kind: dnd.SceneKindNarrative, Title: "Arrival", Summary: "The party arrives."}}
+	req := extractionRequest()
+	if _, err := newExtractor(t, client).Extract(context.Background(), req); err != nil {
+		t.Fatalf("Extract() error = %v, want nil", err)
+	}
+	for _, slot := range []string{"players", "party", "glossary"} {
+		if got := string(client.requests[0].Inputs[slot].Content); got != " " {
+			t.Fatalf("empty %s input = %q, want explicit placeholder", slot, got)
+		}
+	}
+
+	req.References = contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		"players":  {Slot: contracts.ReferenceSlot{Name: "players"}, Items: []contracts.ReferenceItem{{SlotName: "players", Content: []byte("Dana: Mira")}}},
+		"party":    {Slot: contracts.ReferenceSlot{Name: "party"}, Items: []contracts.ReferenceItem{{SlotName: "party", Content: []byte("Mira: ranger")}}},
+		"glossary": {Slot: contracts.ReferenceSlot{Name: "glossary"}, Items: []contracts.ReferenceItem{{SlotName: "glossary", Content: []byte("Greencloak: local title")}}},
+	}}
+	if _, err := newExtractor(t, client).Extract(context.Background(), req); err != nil {
+		t.Fatalf("Extract() error = %v, want nil", err)
+	}
+	inputs := client.requests[1].Inputs
+	if string(inputs["players"].Content) != "Dana: Mira" || string(inputs["party"].Content) != "Mira: ranger" || string(inputs["glossary"].Content) != "Greencloak: local title" {
+		t.Fatalf("reference inputs = %#v", inputs)
+	}
+	if strings.Contains(string(inputs["transcript"].Content), "local title") {
+		t.Fatal("transcript input contains reference content")
+	}
+}
+
+func TestExtractPreservesTheModelKindWithoutRepair(t *testing.T) {
+	client := &fakeSceneDescriptionsLLMClient{response: extractionResponse{
+		Kind: dnd.SceneKind("unrecognized"), Title: "  Untitled scene ", Summary: " Summary ",
+	}}
+	result, err := newExtractor(t, client).Extract(context.Background(), extractionRequest())
+	if err != nil {
+		t.Fatalf("Extract() error = %v, want nil", err)
+	}
+	scene := result.Value.Scenes[0]
+	if scene.Kind != dnd.SceneKind("unrecognized") || scene.Title != "Untitled scene" || scene.Summary != "Summary" {
+		t.Fatalf("scene = %#v, want model kind preserved and textual fields trimmed", scene)
+	}
+}
+
+func TestExtractValidatesRequestsAndSurfacesProviderFailures(t *testing.T) {
+	request := extractionRequest()
+	extractor := newExtractor(t, &fakeSceneDescriptionsLLMClient{response: extractionResponse{Kind: dnd.SceneKindMeta, Title: "Table talk", Summary: "The group discusses rules."}})
+	for _, test := range []struct {
+		name string
+		req  contracts.TypedExtractionRequest
+	}{
+		{name: "nil source", req: func() contracts.TypedExtractionRequest { r := request; r.Source = nil; return r }()},
+		{name: "nil chunk", req: func() contracts.TypedExtractionRequest { r := request; r.Chunk = nil; return r }()},
+		{name: "empty chunk", req: emptyChunkRequest(request)},
+		{name: "mismatched source input", req: mismatchedSourceInputRequest(request)},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := extractor.Extract(context.Background(), test.req); err == nil || !strings.Contains(err.Error(), "dnd scene descriptions") {
+				t.Fatalf("Extract() error = %v, want contextual validation error", err)
+			}
+		})
+	}
+	canceled, cancel := context.WithCancel(context.Background())
+	cancel()
+	if _, err := extractor.Extract(canceled, request); err == nil || !strings.Contains(err.Error(), "context") {
+		t.Fatalf("canceled Extract() error = %v, want context error", err)
+	}
+	_, err := newExtractor(t, &fakeSceneDescriptionsLLMClient{err: errors.New("provider unavailable")}).Extract(context.Background(), request)
+	if err == nil || !strings.Contains(err.Error(), "dnd scene descriptions") || !strings.Contains(err.Error(), "provider unavailable") {
+		t.Fatalf("provider Extract() error = %v, want contextual provider error", err)
+	}
+}
+
+func TestMapResponseUsesOnlyChunkIdentityAndRange(t *testing.T) {
+	chunk := &source.Chunk{ID: "source:chunk:7", Ref: source.SourceRef{SourceID: "source", StartUnitID: 12, EndUnitID: 14}}
+	got := mapResponse(extractionResponse{Kind: dnd.SceneKindRecap, Title: " Recap ", Summary: " Summary "}, chunk)
+	want := dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{
+		ID: "source:chunk:7", SourceRef: source.SourceRef{SourceID: "source", StartUnitID: 12, EndUnitID: 14},
+		Kind: dnd.SceneKindRecap, Title: "Recap", Summary: "Summary",
+	}}}
+	if !reflect.DeepEqual(got, want) {
+		t.Fatalf("mapResponse() = %#v, want %#v", got, want)
+	}
+}

internal/modules/dnd/extract/scenedescriptions/model.go

@@ -0,0 +1,9 @@
+package scenedescriptions
+
+import "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+
+type extractionResponse struct {
+	Kind    dnd.SceneKind `json:"kind"`
+	Title   string        `json:"title"`
+	Summary string        `json:"summary"`
+}

internal/modules/dnd/extract/scenedescriptions/registry_test.go

@@ -0,0 +1,92 @@
+package scenedescriptions
+
+import (
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+func TestNewRequiresLLMClientAndRejectsAmbiguousReferences(t *testing.T) {
+	if _, err := New(nil, Options{}); err == nil || !strings.Contains(err.Error(), "LLM client") {
+		t.Fatalf("New(nil) error = %v, want dependency error", err)
+	}
+	if _, err := New(&fakeSceneDescriptionsLLMClient{}, Options{}, contracts.ReferenceSet{}, contracts.ReferenceSet{}); err == nil || !strings.Contains(err.Error(), "at most one reference set") {
+		t.Fatalf("New() error = %v, want reference-set error", err)
+	}
+	if got := newExtractor(t, &fakeSceneDescriptionsLLMClient{}).Key(); got != Key {
+		t.Fatalf("Key() = %q, want %q", got, Key)
+	}
+}
+
+func TestModuleSpecAndReferenceSlots(t *testing.T) {
+	want := pipeline.ModuleSpec{
+		Key: Key, Stage: pipeline.StageExtract, Requires: []string{"chunks", "source.transcript"}, Provides: []string{"dnd.scene_descriptions"}, ArtifactKind: dnd.SceneDescriptionListKind,
+		ReferenceSlots: []contracts.ReferenceSlot{
+			{Name: "glossary", Description: "Optional campaign glossary reference material used only to disambiguate scene descriptions.", AcceptedMediaTypes: []string{"application/json", "application/x-yaml", "application/yaml", "text/markdown", "text/plain"}},
+			{Name: "party", Description: "Optional party roster reference material used only to disambiguate scene descriptions.", AcceptedMediaTypes: []string{"application/json", "application/x-yaml", "application/yaml", "text/markdown", "text/plain"}},
+			{Name: "players", Description: "Optional player list reference material used only to disambiguate scene descriptions.", AcceptedMediaTypes: []string{"application/json", "application/x-yaml", "application/yaml", "text/markdown", "text/plain"}},
+		},
+	}
+	if got := ModuleSpec(); !reflect.DeepEqual(got, want) {
+		t.Fatalf("ModuleSpec() = %#v, want %#v", got, want)
+	}
+	got := ModuleSpec()
+	got.Requires[0] = "changed"
+	got.Provides[0] = "changed"
+	got.ReferenceSlots[0].AcceptedMediaTypes[0] = "changed"
+	if !reflect.DeepEqual(ModuleSpec(), want) {
+		t.Fatal("ModuleSpec() returned mutable shared slices")
+	}
+	if slots := newExtractor(t, &fakeSceneDescriptionsLLMClient{}).ReferenceSlots(); !reflect.DeepEqual(slots, want.ReferenceSlots) {
+		t.Fatalf("ReferenceSlots() = %#v, want %#v", slots, want.ReferenceSlots)
+	}
+}
+
+func TestRegisterStoresTypedModuleSpecAndOptionsAreStrict(t *testing.T) {
+	registry := pipeline.NewExtractorRegistry()
+	if err := Register(registry); err != nil {
+		t.Fatalf("Register() error = %v, want nil", err)
+	}
+	if got, ok := registry.Spec(Key); !ok || !reflect.DeepEqual(got, ModuleSpec()) {
+		t.Fatalf("registry.Spec(%q) = %#v, present = %t", Key, got, ok)
+	}
+	if err := Register(nil); err == nil || !strings.Contains(err.Error(), "extractor registry") {
+		t.Fatalf("Register(nil) error = %v, want registry error", err)
+	}
+	if _, err := DecodeOptions(map[string]any{"unexpected": true}); err == nil || !strings.Contains(err.Error(), "unknown option") {
+		t.Fatalf("DecodeOptions() error = %v, want strict options error", err)
+	}
+}
+
+func TestExtractorMetadataAndCheckpointIdentity(t *testing.T) {
+	metadata := newExtractor(t, &fakeSceneDescriptionsLLMClient{}).ManifestMetadata()
+	for key, want := range map[string]string{
+		"prompt_id": PromptID, "prompt_version": SchemaVersion,
+		"response_schema_key": string(ResponseSchemaKey), "response_schema_id": ResponseSchemaID,
+		"response_schema_name": ResponseSchemaName, "response_schema_version": SchemaVersion,
+		"mapping_policy": mappingPolicy,
+	} {
+		if metadata[key] != want {
+			t.Fatalf("metadata[%q] = %#v, want %q", key, metadata[key], want)
+		}
+	}
+	for _, key := range []string{"prompt_sha256", "response_schema_sha256"} {
+		if value, ok := metadata[key].(string); !ok || !strings.HasPrefix(value, "sha256:") {
+			t.Fatalf("metadata[%q] = %#v, want hash", key, metadata[key])
+		}
+	}
+	want := map[string]string{"prompt": metadata["prompt_sha256"].(string), "response_schema": metadata["response_schema_sha256"].(string), "mapping_policy": mappingPolicy}
+	fingerprints := newExtractor(t, &fakeSceneDescriptionsLLMClient{}).CheckpointFingerprints()
+	if len(fingerprints) != len(want) {
+		t.Fatalf("CheckpointFingerprints() = %#v, want %d entries", fingerprints, len(want))
+	}
+	for _, fingerprint := range fingerprints {
+		if fingerprint.Value != want[fingerprint.Name] {
+			t.Fatalf("fingerprint %q = %q, want %#v", fingerprint.Name, fingerprint.Value, want)
+		}
+	}
+}

internal/modules/dnd/extract/scenedescriptions/schema.go

@@ -0,0 +1,22 @@
+package scenedescriptions
+
+import "gitea.maximumdirect.net/eric/notarius/internal/framework/llm"
+
+const (
+	PromptID           = "dnd.scene_descriptions"
+	ResponseSchemaKey  = llm.ResponseSchemaKey("dnd_scene_descriptions_llm")
+	ResponseSchemaID   = "notarius.dnd.scene_descriptions.llm"
+	ResponseSchemaName = "notarius_dnd_scene_descriptions_llm_v1"
+	SchemaVersion      = "v1"
+	mappingPolicy      = "dnd.scene_descriptions.mapping.v1"
+)
+
+func loadResponseSchema() (llm.ResponseSchema, error) {
+	return llm.LoadResponseSchema(embeddedAssets, llm.ResponseSchemaDefinition{
+		Key:       ResponseSchemaKey,
+		ID:        ResponseSchemaID,
+		Version:   SchemaVersion,
+		Name:      ResponseSchemaName,
+		AssetPath: "assets/schemas/dnd_scene_descriptions_llm.v1.json",
+	})
+}

internal/modules/dnd/extract/scenedescriptions/schema_test.go

@@ -0,0 +1,92 @@
+package scenedescriptions
+
+import (
+	"bytes"
+	"encoding/json"
+	"strings"
+	"testing"
+
+	"github.com/santhosh-tekuri/jsonschema/v6"
+)
+
+func TestLoadResponseSchemaUsesStrictPrivateSceneDescriptionContract(t *testing.T) {
+	schema, err := loadResponseSchema()
+	if err != nil {
+		t.Fatalf("loadResponseSchema() error = %v, want nil", err)
+	}
+	if schema.Key != ResponseSchemaKey || schema.ID != ResponseSchemaID || schema.Version != SchemaVersion || schema.Name != ResponseSchemaName || !strings.HasPrefix(schema.SHA256, "sha256:") || !json.Valid(schema.JSONSchema) {
+		t.Fatalf("schema = %#v, want private scene-description schema identity", schema)
+	}
+	valid := map[string]any{"kind": "narrative", "title": "Arrival", "summary": "The party enters the tower."}
+	if err := validateJSONSchema(t, valid, schema.JSONSchema); err != nil {
+		t.Fatalf("valid private response rejected: %v", err)
+	}
+	for _, test := range []struct {
+		name     string
+		response map[string]any
+		valid    bool
+	}{
+		{name: "missing required field", response: map[string]any{"kind": "combat", "title": "Ambush"}},
+		{name: "unknown framework field", response: map[string]any{"kind": "combat", "title": "Ambush", "summary": "Bandits strike.", "id": "assigned-later"}},
+		{name: "unknown application field", response: map[string]any{"kind": "combat", "title": "Ambush", "summary": "Bandits strike.", "source_ref": map[string]any{}}},
+		{name: "collection is not allowed", response: map[string]any{"kind": "combat", "title": "Ambush", "summary": "Bandits strike.", "scenes": []any{}}},
+		{name: "unsupported kind", response: map[string]any{"kind": "interlude", "title": "Ambush", "summary": "Bandits strike."}},
+		{name: "wrong field type", response: map[string]any{"kind": "combat", "title": 7, "summary": "Bandits strike."}},
+		{name: "blank title", response: map[string]any{"kind": "combat", "title": "", "summary": "Bandits strike."}},
+		{name: "blank summary", response: map[string]any{"kind": "combat", "title": "Ambush", "summary": ""}},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			err := validateJSONSchema(t, test.response, schema.JSONSchema)
+			if (err == nil) != test.valid {
+				t.Fatalf("validateJSONSchema() error = %v, want valid=%t", err, test.valid)
+			}
+		})
+	}
+	if err := validateJSONSchemaContent([]byte(`{"kind":`), schema.JSONSchema); err == nil {
+		t.Fatal("validateJSONSchemaContent() error = nil, want malformed JSON rejected")
+	}
+}
+
+func TestResponseSchemaIsMutationSafeAndDiagnosticsRedactContent(t *testing.T) {
+	first, err := loadResponseSchema()
+	if err != nil {
+		t.Fatal(err)
+	}
+	first.JSONSchema[0] = '['
+	second, err := loadResponseSchema()
+	if err != nil || !json.Valid(second.JSONSchema) || bytes.Equal(first.JSONSchema, second.JSONSchema) {
+		t.Fatalf("second schema = %s, %v; want defensive copy", second.JSONSchema, err)
+	}
+	if diagnostics := second.DiagnosticsMap(); diagnostics["json_schema"] != nil {
+		t.Fatalf("schema diagnostics included raw content: %#v", diagnostics)
+	}
+}
+
+func validateJSONSchema(t *testing.T, instance map[string]any, schemaContent []byte) error {
+	t.Helper()
+	content, err := json.Marshal(instance)
+	if err != nil {
+		t.Fatal(err)
+	}
+	return validateJSONSchemaContent(content, schemaContent)
+}
+
+func validateJSONSchemaContent(instanceContent, schemaContent []byte) error {
+	instance, err := jsonschema.UnmarshalJSON(bytes.NewReader(instanceContent))
+	if err != nil {
+		return err
+	}
+	schemaDocument, err := jsonschema.UnmarshalJSON(bytes.NewReader(schemaContent))
+	if err != nil {
+		return err
+	}
+	compiler := jsonschema.NewCompiler()
+	if err := compiler.AddResource("schema.json", schemaDocument); err != nil {
+		return err
+	}
+	compiled, err := compiler.Compile("schema.json")
+	if err != nil {
+		return err
+	}
+	return compiled.Validate(instance)
+}

internal/modules/dnd/extract/scenedescriptions/scriptorium_assets.go

@@ -0,0 +1,51 @@
+package scenedescriptions
+
+import (
+	"fmt"
+	"sync"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/llm"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/promptfs"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
+)
+
+const scriptoriumPromptRoot = "assets/prompts"
+
+var promptAssetManifest = shared.PromptAssetManifest{
+	ModuleDir: "dnd.scene_descriptions",
+	ModuleFiles: []promptfs.ModulePromptFile{
+		{Name: "dnd.scene_descriptions.yaml", Path: "assets/prompts/dnd.scene_descriptions.yaml"},
+		{Name: "task.md", Path: "assets/prompts/task.md"},
+		{Name: "instructions.md", Path: "assets/prompts/instructions.md"},
+	},
+	SharedFiles: []string{
+		"common-dnd-system.md",
+		"common-dnd-identity.md",
+		"common-dnd-references.md",
+		"common-dnd-transcript.md",
+	},
+}
+
+func RegisterPromptAssets(registry *llm.AssetRegistry) error {
+	promptFS, err := promptAssetManifest.PromptFS(embeddedAssets)
+	if err != nil {
+		return fmt.Errorf("prepare scene-description prompt assets: %w", err)
+	}
+	if err := registry.RegisterPromptFS(promptFS, scriptoriumPromptRoot); err != nil {
+		return err
+	}
+	return registry.RegisterSchemaFS(embeddedAssets, "assets/schemas")
+}
+
+func scriptoriumPromptMetadata() (string, error) {
+	scriptoriumPromptHashOnce.Do(func() {
+		scriptoriumPromptHash, scriptoriumPromptHashErr = promptAssetManifest.Hash(embeddedAssets)
+	})
+	return scriptoriumPromptHash, scriptoriumPromptHashErr
+}
+
+var (
+	scriptoriumPromptHashOnce sync.Once
+	scriptoriumPromptHash     string
+	scriptoriumPromptHashErr  error
+)

internal/modules/dnd/extract/scenedescriptions/scriptorium_assets_test.go

@@ -0,0 +1,106 @@
+package scenedescriptions
+
+import (
+	"context"
+	"encoding/json"
+	"reflect"
+	"strings"
+	"testing"
+	"time"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/llm"
+	"gitea.maximumdirect.net/eric/scriptorium"
+)
+
+func TestRegisterPromptAssetsPreparesOrderedSceneDescriptionPrompt(t *testing.T) {
+	registry := llm.NewAssetRegistry()
+	if err := RegisterPromptAssets(registry); err != nil {
+		t.Fatalf("RegisterPromptAssets() error = %v, want nil", err)
+	}
+	options, err := registry.ScriptoriumOptions()
+	if err != nil {
+		t.Fatalf("ScriptoriumOptions() error = %v, want nil", err)
+	}
+	options = append(options, scriptorium.WithProfiles(scriptorium.OpenAICompatibleProfile(scriptorium.OpenAICompatibleProfileConfig{
+		ID: "scene-description-test-profile", Endpoint: "http://127.0.0.1:1/v1", Model: "scene-description-test-model",
+	})))
+	engine, err := scriptorium.NewEngine(scriptorium.Config{Timeout: time.Second}, options...)
+	if err != nil {
+		t.Fatalf("NewEngine() error = %v, want nil", err)
+	}
+	prepared, err := engine.Prepare(context.Background(), scriptorium.RunRequest{
+		PromptID: PromptID, PromptVersion: SchemaVersion, ProfileID: "scene-description-test-profile",
+		Inputs: map[string]scriptorium.ArtifactRef{
+			"transcript": scriptorium.InlineWithURI("file:///session.json", `{"units":[1]}`),
+			"players":    scriptorium.Inline("Dana: Mira"),
+			"party":      scriptorium.Inline("Mira: ranger"),
+			"glossary":   scriptorium.Inline("Greencloak: title"),
+		},
+	})
+	if err != nil {
+		t.Fatalf("Prepare() error = %v, want nil", err)
+	}
+	if prepared.PromptID != PromptID || prepared.OutputContract.SchemaPath != "dnd_scene_descriptions_llm.v1.json" {
+		t.Fatalf("prepared prompt = %#v, want scene-description prompt identity and schema wiring", prepared)
+	}
+	if got, want := messageRoles(prepared.Messages), []string{"system", "user", "user", "user", "user", "user"}; !reflect.DeepEqual(got, want) {
+		t.Fatalf("message roles = %#v, want %#v", got, want)
+	}
+	for _, index := range []int{1, 2, 4} {
+		if prepared.Messages[index].CacheControl == nil || prepared.Messages[index].CacheControl.Type != scriptorium.CacheControlEphemeral {
+			t.Fatalf("message %d cache control = %#v, want ephemeral", index, prepared.Messages[index].CacheControl)
+		}
+	}
+	for _, index := range []int{0, 3, 5} {
+		if prepared.Messages[index].CacheControl != nil {
+			t.Fatalf("message %d cache control = %#v, want nil", index, prepared.Messages[index].CacheControl)
+		}
+	}
+	if !strings.Contains(prepared.Messages[0].Content, "Dungeons & Dragons") {
+		t.Fatalf("first message does not use shared D&D system asset: %q", prepared.Messages[0].Content)
+	}
+	for index, want := range []string{"Dana: Mira", "Mira: ranger", "Greencloak: title"} {
+		if !strings.Contains(prepared.Messages[2].Content, want) {
+			t.Fatalf("reference %d not rendered in shared reference message: %q", index, prepared.Messages[2].Content)
+		}
+	}
+	if strings.Contains(prepared.Messages[1].Content, `{"units":[1]}`) || strings.Contains(prepared.Messages[2].Content, `{"units":[1]}`) {
+		t.Fatal("transcript rendered before its final message")
+	}
+	if !strings.Contains(prepared.Messages[5].Content, `{"units":[1]}`) {
+		t.Fatalf("final message does not render transcript: %q", prepared.Messages[5].Content)
+	}
+	transcriptMessages := 0
+	for _, message := range prepared.Messages {
+		if strings.Contains(message.Content, `{"units":[1]}`) {
+			transcriptMessages++
+		}
+	}
+	if transcriptMessages != 1 {
+		t.Fatalf("raw transcript rendered in %d messages, want exactly one", transcriptMessages)
+	}
+}
+
+func TestPromptMetadataAndDiagnosticsDoNotContainRawAssets(t *testing.T) {
+	hash, err := scriptoriumPromptMetadata()
+	if err != nil || !strings.HasPrefix(hash, "sha256:") {
+		t.Fatalf("scriptoriumPromptMetadata() = %q, %v; want hash", hash, err)
+	}
+	payload, err := json.Marshal(newExtractor(t, &fakeSceneDescriptionsLLMClient{}).ManifestMetadata())
+	if err != nil {
+		t.Fatal(err)
+	}
+	for _, forbidden := range []string{"Choose exactly one kind", "common-dnd-system", "dnd_scene_descriptions_llm.v1.json"} {
+		if strings.Contains(string(payload), forbidden) {
+			t.Fatalf("metadata leaked raw prompt/schema content %q: %s", forbidden, payload)
+		}
+	}
+}
+
+func messageRoles(messages []scriptorium.RenderedMessage) []string {
+	roles := make([]string, len(messages))
+	for i, message := range messages {
+		roles[i] = string(message.Role)
+	}
+	return roles
+}

internal/modules/dnd/extract/scenedescriptions/test_helpers_test.go

@@ -0,0 +1,102 @@
+package scenedescriptions
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+)
+
+func extractionRequest() contracts.TypedExtractionRequest {
+	doc := sourceDocument()
+	chunk := &source.Chunk{
+		ID:        "session-alpha:chunk:0",
+		SourceID:  doc.ID,
+		Index:     0,
+		Ref:       source.SourceRef{SourceID: doc.ID, StartUnitID: 1, EndUnitID: 3},
+		Content:   []byte(`{"units":[1,2,3]}`),
+		MediaType: "application/json",
+		Units:     append([]source.SourceUnit(nil), doc.Units...),
+		Metadata:  map[string]any{"ignored": "chunk metadata"},
+	}
+	return contracts.TypedExtractionRequest{
+		Source:      doc,
+		Chunk:       chunk,
+		SourceInput: contracts.NewLLMInputMaterial("source", chunk.MediaType, chunk.Content, "sha256:chunk", "file:///session-alpha.json"),
+		SessionID:   "session-123",
+		LLMProfile:  "profile-scene-descriptions",
+	}
+}
+
+func sourceDocument() *source.SourceDocument {
+	return &source.SourceDocument{
+		ID:     "session-alpha",
+		Kind:   "transcript",
+		Format: "application/vnd.seriatim.minimal+json",
+		Digest: "sha256:test",
+		Units: []source.SourceUnit{
+			{ID: 1, Kind: "transcript_segment", Text: "Mira Thorn enters the ruined watchtower."},
+			{ID: 2, Kind: "transcript_segment", Text: "Bandits attack from the upper floor."},
+			{ID: 3, Kind: "transcript_segment", Text: "The party drives them back."},
+		},
+	}
+}
+
+func newExtractor(t *testing.T, client contracts.StructuredLLMClient, references ...contracts.ReferenceSet) *Extractor {
+	t.Helper()
+	extractor, err := New(client, Options{}, references...)
+	if err != nil {
+		t.Fatalf("New() error = %v, want nil", err)
+	}
+	return extractor
+}
+
+func emptyChunkRequest(req contracts.TypedExtractionRequest) contracts.TypedExtractionRequest {
+	req.Chunk = &source.Chunk{ID: req.Chunk.ID, SourceID: req.Chunk.SourceID, Index: req.Chunk.Index}
+	return req
+}
+
+func mismatchedSourceInputRequest(req contracts.TypedExtractionRequest) contracts.TypedExtractionRequest {
+	req.SourceInput = contracts.NewLLMInputMaterial("source", "application/json", []byte(`{"different":true}`), "sha256:other", "file:///other.json")
+	return req
+}
+
+func cloneStructuredCompletionRequest(req contracts.StructuredCompletionRequest) contracts.StructuredCompletionRequest {
+	req.Inputs = req.Inputs.Clone()
+	if len(req.Vars) == 0 {
+		req.Vars = nil
+		return req
+	}
+	vars := make(map[string]any, len(req.Vars))
+	for key, value := range req.Vars {
+		vars[key] = value
+	}
+	req.Vars = vars
+	return req
+}
+
+type fakeSceneDescriptionsLLMClient struct {
+	response extractionResponse
+	err      error
+	requests []contracts.StructuredCompletionRequest
+}
+
+func (client *fakeSceneDescriptionsLLMClient) CompleteStructured(_ context.Context, req contracts.StructuredCompletionRequest, out any) (contracts.StructuredCompletionResponse, error) {
+	client.requests = append(client.requests, cloneStructuredCompletionRequest(req))
+	if client.err != nil {
+		return contracts.StructuredCompletionResponse{}, client.err
+	}
+	target, ok := out.(*extractionResponse)
+	if !ok {
+		return contracts.StructuredCompletionResponse{}, errors.New("unexpected output target")
+	}
+	*target = client.response
+	content, err := json.Marshal(client.response)
+	if err != nil {
+		return contracts.StructuredCompletionResponse{}, err
+	}
+	return contracts.StructuredCompletionResponse{Content: content}, nil
+}

internal/modules/dnd/normalize/npcinteractions/normalizer.go

@@ -0,0 +1,307 @@
+// Package npcinteractions normalizes merged D&D NPC interaction candidates.
+package npcinteractions
+
+import (
+	"context"
+	"fmt"
+	"sort"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	interactionmodel "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcinteractions"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
+	npcregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/registry"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
+)
+
+const (
+	Key                 = "dnd/npc-interactions"
+	normalizationPolicy = "dnd.npc_interactions.normalize.v2"
+	NormalizationPolicy = normalizationPolicy
+
+	ReasonCodeNameCanonicalized     = "npc_interaction_name_canonicalized"
+	ReasonCodeSourceRefsNormalized  = "source_references_normalized"
+	ReasonCodeInteractionsReordered = "npc_interactions_reordered"
+	ReasonCodeDuplicateCollapsed    = "duplicate_npc_interaction_collapsed"
+	ReasonCodeWarningsOmitted       = "npc_interaction_normalization_warnings_omitted"
+)
+
+const (
+	NPCRegistryReferenceSlot = npcregistry.ReferenceSlot
+	NPCRegistryMaxBytes      = npcregistry.MaxBytes
+)
+
+var requiredCapabilities = []string{"merged"}
+var providedCapabilities = []string{"normalized"}
+
+var referenceSlotDescriptions = shared.ReferenceSlotDescriptions{
+	Glossary: "Optional campaign glossary reference material used only for interaction disambiguation.",
+	Party:    "Optional party roster reference material used only for interaction disambiguation.",
+	Players:  "Optional player list reference material used only for interaction disambiguation.",
+	Roster:   "Deprecated alias for party roster reference material used only for interaction disambiguation.",
+}
+
+var _ contracts.Normalizer[dnd.NPCInteractionList] = (*Normalizer)(nil)
+var _ contracts.ManifestMetadataProvider = (*Normalizer)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Normalizer)(nil)
+
+type Options struct{}
+
+type Normalizer struct {
+	npcResolver *npcregistry.Resolver
+}
+
+func New(_ Options, references ...contracts.ReferenceSet) (*Normalizer, error) {
+	if len(references) > 1 {
+		return nil, normalizerErrorf("at most one reference set may be supplied")
+	}
+	var referenceSet contracts.ReferenceSet
+	if len(references) == 1 {
+		referenceSet = references[0]
+	}
+	resolver, err := npcregistry.NewResolver(referenceSet)
+	if err != nil {
+		return nil, normalizerErrorf("prepare NPC registry: %w", err)
+	}
+	return &Normalizer{npcResolver: resolver}, nil
+}
+
+func (n *Normalizer) Key() string { return Key }
+
+func (n *Normalizer) ReferenceSlots() []contracts.ReferenceSlot { return referenceSlots() }
+
+func (n *Normalizer) ManifestMetadata() map[string]any {
+	if n == nil || n.npcResolver == nil {
+		return nil
+	}
+	metadata := map[string]any{
+		"normalization_policy": normalizationPolicy,
+		"identity_policy":      identity.Policy,
+	}
+	seeded := n.npcResolver.Seeded()
+	if seeded.Bound() {
+		metadata["npc_registry_digest"] = seeded.Digest()
+		metadata["npc_count"] = seeded.Count()
+	}
+	return metadata
+}
+
+func (n *Normalizer) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	if n == nil || n.npcResolver == nil {
+		return nil
+	}
+	return []pipeline.CheckpointFingerprint{
+		{Name: "normalization_policy", Value: normalizationPolicy},
+		{Name: "identity_policy", Value: identity.Policy},
+		{Name: "npc_registry", Value: n.npcResolver.Seeded().ProjectionDigest()},
+	}
+}
+
+func (n *Normalizer) Normalize(ctx context.Context, req contracts.TypedNormalizeRequest[dnd.NPCInteractionList]) (contracts.TypedNormalizeResult[dnd.NPCInteractionList], error) {
+	if n == nil || n.npcResolver == nil {
+		return contracts.TypedNormalizeResult[dnd.NPCInteractionList]{}, normalizerErrorf("normalizer must not be nil")
+	}
+	if ctx == nil {
+		return contracts.TypedNormalizeResult[dnd.NPCInteractionList]{}, normalizerErrorf("context must not be nil")
+	}
+	if err := ctx.Err(); err != nil {
+		return contracts.TypedNormalizeResult[dnd.NPCInteractionList]{}, normalizerErrorf("context error before normalize: %w", err)
+	}
+
+	registry, err := n.npcResolver.Resolve(req.References)
+	if err != nil {
+		return contracts.TypedNormalizeResult[dnd.NPCInteractionList]{}, normalizerErrorf("resolve NPC registry: %w", err)
+	}
+	if !registry.Bound() {
+		return contracts.TypedNormalizeResult[dnd.NPCInteractionList]{}, normalizerErrorf("NPC registry reference is required")
+	}
+	value, warnings := normalizeList(req.MergeOutput.Value, req.Source, registry)
+	return contracts.TypedNormalizeResult[dnd.NPCInteractionList]{Value: value, Warnings: warnings}, nil
+}
+
+type normalizedRecord struct {
+	interaction dnd.NPCInteraction
+	inputIndex  int
+}
+
+type nameCanonicalization struct {
+	from string
+	to   string
+}
+
+func normalizeList(input dnd.NPCInteractionList, doc *source.SourceDocument, registry *npcregistry.Registry) (dnd.NPCInteractionList, []contracts.Warning) {
+	if input.Interactions == nil {
+		return dnd.NPCInteractionList{}, nil
+	}
+
+	records := make([]normalizedRecord, len(input.Interactions))
+	warnings := make([]contracts.Warning, 0)
+	for index, inputInteraction := range input.Interactions {
+		interaction, nameChange, refsChanged := normalizeInteraction(inputInteraction, doc, registry)
+		records[index] = normalizedRecord{interaction: interaction, inputIndex: index}
+		if nameChange != nil {
+			warnings = append(warnings, contracts.Warning{
+				Scope:      interactionScope(index),
+				ReasonCode: ReasonCodeNameCanonicalized,
+				Message: fmt.Sprintf("input index %d: NPC name canonicalized from %s to %s",
+					index, diagnostics.Quote(nameChange.from), diagnostics.Quote(nameChange.to)),
+			})
+		}
+		if refsChanged {
+			warnings = append(warnings, contracts.Warning{
+				Scope:      interactionScope(index),
+				ReasonCode: ReasonCodeSourceRefsNormalized,
+				Message: fmt.Sprintf("input index %d: source references normalized (original count %d, final count %d)",
+					index, len(inputInteraction.SourceRefs), len(interaction.SourceRefs)),
+			})
+		}
+	}
+
+	sort.SliceStable(records, func(left, right int) bool {
+		return interactionmodel.Less(doc, records[left].interaction, records[right].interaction)
+	})
+	for position, record := range records {
+		if position == record.inputIndex {
+			continue
+		}
+		warnings = append(warnings, contracts.Warning{
+			Scope:      interactionScope(record.inputIndex),
+			ReasonCode: ReasonCodeInteractionsReordered,
+			Message:    fmt.Sprintf("input index %d moved to normalized position %d by source chronology", record.inputIndex, position),
+		})
+	}
+
+	output, duplicateWarnings := collapseDuplicates(records, doc)
+	warnings = append(warnings, duplicateWarnings...)
+	return dnd.NPCInteractionList{Interactions: output},
+		diagnostics.LimitWarnings(warnings, "npc_interactions", ReasonCodeWarningsOmitted)
+}
+
+func normalizeInteraction(input dnd.NPCInteraction, doc *source.SourceDocument, registry *npcregistry.Registry) (dnd.NPCInteraction, *nameCanonicalization, bool) {
+	output := cloneInteraction(input)
+	if canonical, ok := registry.Lookup(identity.NormalizeDisplay(input.Name)); ok {
+		output.Name = canonical.Name
+	}
+	var nameChange *nameCanonicalization
+	if input.Name != output.Name {
+		nameChange = &nameCanonicalization{from: input.Name, to: output.Name}
+	}
+	output.SourceRefs = interactionmodel.CanonicalizeSourceRefs(doc, input.SourceRefs)
+	return output, nameChange, !interactionmodel.SourceRefsEqual(input.SourceRefs, output.SourceRefs)
+}
+
+func cloneInteraction(input dnd.NPCInteraction) dnd.NPCInteraction {
+	output := input
+	if input.SourceRefs != nil {
+		output.SourceRefs = append([]source.SourceRef(nil), input.SourceRefs...)
+	}
+	return output
+}
+
+type duplicateGroup struct {
+	retainedIndex int
+	removed       []int
+}
+
+func collapseDuplicates(records []normalizedRecord, doc *source.SourceDocument) ([]dnd.NPCInteraction, []contracts.Warning) {
+	if len(records) == 0 {
+		return make([]dnd.NPCInteraction, 0), nil
+	}
+	keep := make([]bool, len(records))
+	groups := make([]duplicateGroup, 0)
+	groupByKey := make(map[string]int)
+	for index, record := range records {
+		if !interactionmodel.ValidSourceRefs(doc, record.interaction.SourceRefs) {
+			keep[index] = true
+			continue
+		}
+		key := interactionmodel.ExactIdentity(record.interaction)
+		groupIndex, exists := groupByKey[key]
+		if !exists {
+			groupByKey[key] = len(groups)
+			groups = append(groups, duplicateGroup{retainedIndex: record.inputIndex})
+			keep[index] = true
+			continue
+		}
+		groups[groupIndex].removed = append(groups[groupIndex].removed, record.inputIndex)
+	}
+	output := make([]dnd.NPCInteraction, 0, len(records))
+	for index, record := range records {
+		if keep[index] {
+			output = append(output, cloneInteraction(record.interaction))
+		}
+	}
+	warnings := make([]contracts.Warning, 0)
+	for _, group := range groups {
+		if len(group.removed) != 0 {
+			warnings = append(warnings, duplicateWarning(group.retainedIndex, group.removed))
+		}
+	}
+	return output, warnings
+}
+
+func duplicateWarning(retainedIndex int, removed []int) contracts.Warning {
+	issues := make([]string, len(removed))
+	for index, removedIndex := range removed {
+		issues[index] = fmt.Sprintf("removed input index %d", removedIndex)
+	}
+	return contracts.Warning{
+		Scope:      interactionScope(retainedIndex),
+		ReasonCode: ReasonCodeDuplicateCollapsed,
+		Message: diagnostics.Aggregate(
+			fmt.Sprintf("duplicate NPC interaction collapsed; retained input index %d", retainedIndex), issues),
+	}
+}
+
+func interactionScope(index int) string { return fmt.Sprintf("interactions[%d]", index) }
+
+func referenceSlots() []contracts.ReferenceSlot {
+	slots := shared.ReferenceSlots(referenceSlotDescriptions)
+	slots = append(slots, contracts.ReferenceSlot{
+		Name:                  NPCRegistryReferenceSlot,
+		Description:           "Required normalized NPC registry used only for interaction identity grounding, never as interaction evidence.",
+		Required:              true,
+		AcceptedMediaTypes:    []string{"application/json"},
+		AcceptedArtifactKinds: []contracts.ArtifactKind{dnd.NPCListKind},
+		MaxBytes:              NPCRegistryMaxBytes,
+	})
+	sort.Slice(slots, func(left, right int) bool { return slots[left].Name < slots[right].Name })
+	return slots
+}
+
+func ModuleSpec() pipeline.ModuleSpec {
+	return pipeline.ModuleSpec{
+		Key:            Key,
+		Stage:          pipeline.StageNormalize,
+		Requires:       append([]string(nil), requiredCapabilities...),
+		Provides:       append([]string(nil), providedCapabilities...),
+		ArtifactKind:   dnd.NPCInteractionListKind,
+		ReferenceSlots: referenceSlots(),
+	}
+}
+
+func Register(registry *pipeline.NormalizerRegistry) error {
+	return pipeline.RegisterNormalizerBuilder(registry, ModuleSpec(), validateOptions, func(request pipeline.BuildRequest) (contracts.Normalizer[dnd.NPCInteractionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(options, request.References)
+	})
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, normalizerErrorf("%w", err)
+	}
+	return Options{}, nil
+}
+
+func validateOptions(options map[string]any) error { _, err := DecodeOptions(options); return err }
+
+func normalizerErrorf(format string, args ...any) error {
+	return fmt.Errorf("dnd NPC interactions normalizer: "+format, args...)
+}

internal/modules/dnd/normalize/npcinteractions/normalizer_test.go

@@ -0,0 +1,199 @@
+package npcinteractions
+
+import (
+	"context"
+	"reflect"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	npccodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcs"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
+)
+
+func TestNormalizeCanonicalizesAndClones(t *testing.T) {
+	doc := testDocument()
+	normalizer, err := New(Options{}, npcReferences(t))
+	if err != nil {
+		t.Fatalf("New() error = %v", err)
+	}
+	input := dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{{
+		Name: "  áRIA  ", Kind: dnd.NPCInteractionKindDialogue,
+		SourceRefs: []source.SourceRef{{SourceID: doc.ID, StartUnitID: 30, EndUnitID: 30}, {SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}, {SourceID: doc.ID, StartUnitID: 30, EndUnitID: 30}},
+	}}}
+	original := append([]source.SourceRef(nil), input.Interactions[0].SourceRefs...)
+	result, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{Source: doc, MergeOutput: contracts.MergeArtifact[dnd.NPCInteractionList]{Value: input}})
+	if err != nil {
+		t.Fatalf("Normalize() error = %v", err)
+	}
+	got := result.Value.Interactions[0]
+	if got.Name != "Ária" || !reflect.DeepEqual(got.SourceRefs, []source.SourceRef{{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10}, {SourceID: doc.ID, StartUnitID: 30, EndUnitID: 30}}) {
+		t.Fatalf("normalized interaction = %#v", got)
+	}
+	if !hasWarning(result.Warnings, ReasonCodeNameCanonicalized) || !hasWarning(result.Warnings, ReasonCodeSourceRefsNormalized) {
+		t.Fatalf("warnings = %#v", result.Warnings)
+	}
+	if !reflect.DeepEqual(input.Interactions[0].SourceRefs, original) {
+		t.Fatalf("Normalize() mutated input: %#v", input)
+	}
+	second, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{Source: doc, MergeOutput: contracts.MergeArtifact[dnd.NPCInteractionList]{Value: result.Value}})
+	if err != nil || !reflect.DeepEqual(second.Value, result.Value) || len(second.Warnings) != 0 {
+		t.Fatalf("second normalization = %#v, %v; want idempotent output without warnings", second, err)
+	}
+	result.Value.Interactions[0].SourceRefs[0].StartUnitID = 999
+	if input.Interactions[0].SourceRefs[0].StartUnitID == 999 {
+		t.Fatal("normalized source refs share input storage")
+	}
+}
+
+func TestNormalizeRequiresOperationRegistryAndPreservesEmptyRepresentation(t *testing.T) {
+	normalizer, err := New(Options{})
+	if err != nil {
+		t.Fatalf("New() error = %v", err)
+	}
+	if _, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{}); err == nil {
+		t.Fatal("Normalize() accepted an unbound NPC registry")
+	}
+	for _, input := range []dnd.NPCInteractionList{{}, {Interactions: []dnd.NPCInteraction{}}} {
+		result, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{
+			MergeOutput: contracts.MergeArtifact[dnd.NPCInteractionList]{Value: input}, References: npcReferences(t),
+		})
+		if err != nil || (result.Value.Interactions == nil) != (input.Interactions == nil) {
+			t.Fatalf("Normalize() = %#v, %v for input %#v", result, err, input)
+		}
+	}
+	if metadata := normalizer.ManifestMetadata(); metadata["npc_registry_digest"] != nil || metadata["npc_count"] != nil {
+		t.Fatalf("operation registry leaked into metadata: %#v", metadata)
+	}
+}
+
+func TestNormalizeLeavesUnrecognizedNamesUntouched(t *testing.T) {
+	doc := testDocument()
+	normalizer, err := New(Options{}, npcReferences(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	input := dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{interaction("  Unknown NPC  ", dnd.NPCInteractionKindOther, source.SourceRef{SourceID: doc.ID, StartUnitID: 10, EndUnitID: 10})}}
+	result, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{Source: doc, MergeOutput: contracts.MergeArtifact[dnd.NPCInteractionList]{Value: input}})
+	if err != nil || result.Value.Interactions[0].Name != input.Interactions[0].Name || hasWarning(result.Warnings, ReasonCodeNameCanonicalized) {
+		t.Fatalf("Normalize() = %#v, %v; want untouched unrecognized name", result, err)
+	}
+}
+
+func TestNormalizeOrdersAndCollapsesExactDuplicatesOnly(t *testing.T) {
+	doc := &source.SourceDocument{ID: "session", Units: []source.SourceUnit{{ID: 50}, {ID: 10}, {ID: 90}}}
+	ref := func(unit int) source.SourceRef {
+		return source.SourceRef{SourceID: doc.ID, StartUnitID: unit, EndUnitID: unit}
+	}
+	first := interaction("Ária", dnd.NPCInteractionKindDialogue, ref(50))
+	input := dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{
+		interaction("Borin", dnd.NPCInteractionKindMentioned, ref(90)),
+		first,
+		first,
+		interaction("Ária", dnd.NPCInteractionKindCombatAlly, ref(50)),
+		interaction("Ária", dnd.NPCInteractionKindDialogue, ref(10)),
+		interaction("Ária", dnd.NPCInteractionKindDialogue, ref(999)),
+	}}
+	normalizer, err := New(Options{}, npcReferences(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	result, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{Source: doc, MergeOutput: contracts.MergeArtifact[dnd.NPCInteractionList]{Value: input}})
+	if err != nil {
+		t.Fatal(err)
+	}
+	got := result.Value.Interactions
+	if len(got) != 5 {
+		t.Fatalf("interaction count = %d, want 5: %#v", len(got), got)
+	}
+	if got[0].Kind != dnd.NPCInteractionKindCombatAlly || got[0].SourceRefs[0].StartUnitID != 50 || got[1].SourceRefs[0].StartUnitID != 50 || got[2].SourceRefs[0].StartUnitID != 10 || got[3].SourceRefs[0].StartUnitID != 90 || got[4].SourceRefs[0].StartUnitID != 999 {
+		t.Fatalf("canonical order = %#v", got)
+	}
+	if !hasWarning(result.Warnings, ReasonCodeInteractionsReordered) || !hasWarning(result.Warnings, ReasonCodeDuplicateCollapsed) {
+		t.Fatalf("warnings = %#v", result.Warnings)
+	}
+}
+
+func TestNormalizerContractAndDeterministicWarnings(t *testing.T) {
+	normalizer, err := New(Options{}, npcReferences(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if spec := ModuleSpec(); spec.Key != Key || spec.Stage != pipeline.StageNormalize || spec.ArtifactKind != dnd.NPCInteractionListKind || len(spec.ReferenceSlots) == 0 {
+		t.Fatalf("ModuleSpec() = %#v", spec)
+	}
+	if metadata := normalizer.ManifestMetadata(); metadata["normalization_policy"] != normalizationPolicy || metadata["identity_policy"] != identity.Policy || metadata["npc_registry_digest"] == "" || metadata["npc_count"] != 2 {
+		t.Fatalf("metadata = %#v", metadata)
+	}
+	if fingerprints := normalizer.CheckpointFingerprints(); len(fingerprints) != 3 || fingerprints[2].Name != "npc_registry" || fingerprints[2].Value == "" {
+		t.Fatalf("fingerprints = %#v", fingerprints)
+	}
+	if _, err := DecodeOptions(map[string]any{"unexpected": true}); err == nil {
+		t.Fatal("DecodeOptions() accepted an unknown option")
+	}
+}
+
+func TestNormalizeBoundsWarnings(t *testing.T) {
+	count := diagnostics.MaxWarnings + 5
+	doc := &source.SourceDocument{ID: "session", Units: make([]source.SourceUnit, count)}
+	input := dnd.NPCInteractionList{Interactions: make([]dnd.NPCInteraction, count)}
+	for index := range doc.Units {
+		doc.Units[index].ID = index + 1
+		unitID := count - index
+		input.Interactions[index] = interaction(
+			"Ária",
+			dnd.NPCInteractionKindDialogue,
+			source.SourceRef{SourceID: doc.ID, StartUnitID: unitID, EndUnitID: unitID},
+		)
+	}
+	normalizer, err := New(Options{}, npcReferences(t))
+	if err != nil {
+		t.Fatal(err)
+	}
+	result, err := normalizer.Normalize(context.Background(), contracts.TypedNormalizeRequest[dnd.NPCInteractionList]{
+		Source: doc, MergeOutput: contracts.MergeArtifact[dnd.NPCInteractionList]{Value: input},
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(result.Warnings) != diagnostics.MaxWarnings ||
+		result.Warnings[len(result.Warnings)-1].ReasonCode != ReasonCodeWarningsOmitted {
+		t.Fatalf("warnings = %#v", result.Warnings)
+	}
+}
+
+func interaction(name string, kind dnd.NPCInteractionKind, ref source.SourceRef) dnd.NPCInteraction {
+	return dnd.NPCInteraction{Name: name, Kind: kind, SourceRefs: []source.SourceRef{ref}}
+}
+
+func testDocument() *source.SourceDocument {
+	return &source.SourceDocument{ID: "session", Units: []source.SourceUnit{{ID: 10}, {ID: 20}, {ID: 30}}}
+}
+
+func npcReferences(t *testing.T) contracts.ReferenceSet {
+	t.Helper()
+	value := dnd.NPCList{NPCs: []dnd.NPC{
+		{ID: identity.DeriveID("Ária"), Name: "Ária", SourceRefs: []source.SourceRef{{SourceID: "registry", StartUnitID: 1, EndUnitID: 1}}},
+		{ID: identity.DeriveID("Borin"), Name: "Borin", SourceRefs: []source.SourceRef{{SourceID: "registry", StartUnitID: 1, EndUnitID: 1}}},
+	}}
+	content, err := npccodec.New().Encode(value)
+	if err != nil {
+		t.Fatalf("encode registry: %v", err)
+	}
+	return contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{NPCRegistryReferenceSlot: {
+		Slot:  contracts.ReferenceSlot{Name: NPCRegistryReferenceSlot, Required: true, AcceptedMediaTypes: []string{npccodec.MediaType}, AcceptedArtifactKinds: []contracts.ArtifactKind{dnd.NPCListKind}, MaxBytes: NPCRegistryMaxBytes},
+		Items: []contracts.ReferenceItem{{SlotName: NPCRegistryReferenceSlot, MediaType: npccodec.MediaType, Content: content}},
+	}}}
+}
+
+func hasWarning(warnings []contracts.Warning, reason string) bool {
+	for _, warning := range warnings {
+		if warning.ReasonCode == reason {
+			return true
+		}
+	}
+	return false
+}

internal/modules/dnd/normalize/scenedescriptions/normalizer.go

@@ -0,0 +1,173 @@
+// Package scenedescriptions normalizes merged D&D scene descriptions.
+package scenedescriptions
+
+import (
+	"context"
+	"fmt"
+	"sort"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/shape"
+)
+
+const (
+	Key              = "dnd/scene-descriptions"
+	normalizerPolicy = "dnd.scene_descriptions.normalizer.v1"
+)
+
+var requiredCapabilities = []string{"merged"}
+var providedCapabilities = []string{"normalized"}
+
+var _ contracts.Normalizer[dnd.SceneDescriptionList] = (*Normalizer)(nil)
+var _ contracts.ManifestMetadataProvider = (*Normalizer)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Normalizer)(nil)
+
+type Options struct{}
+type Normalizer struct{}
+
+func New(Options) *Normalizer { return &Normalizer{} }
+
+func (n *Normalizer) Key() string { return Key }
+
+func (n *Normalizer) ReferenceSlots() []contracts.ReferenceSlot { return nil }
+
+func (n *Normalizer) ManifestMetadata() map[string]any {
+	if n == nil {
+		return nil
+	}
+	return map[string]any{"normalizer_policy": normalizerPolicy}
+}
+
+func (n *Normalizer) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	if n == nil {
+		return nil
+	}
+	return []pipeline.CheckpointFingerprint{{Name: "normalizer_policy", Value: normalizerPolicy}}
+}
+
+func (n *Normalizer) Normalize(ctx context.Context, req contracts.TypedNormalizeRequest[dnd.SceneDescriptionList]) (contracts.TypedNormalizeResult[dnd.SceneDescriptionList], error) {
+	if n == nil {
+		return contracts.TypedNormalizeResult[dnd.SceneDescriptionList]{}, normalizerErrorf("normalizer must not be nil")
+	}
+	if ctx == nil {
+		return contracts.TypedNormalizeResult[dnd.SceneDescriptionList]{}, normalizerErrorf("context must not be nil")
+	}
+	if err := ctx.Err(); err != nil {
+		return contracts.TypedNormalizeResult[dnd.SceneDescriptionList]{}, normalizerErrorf("context error before normalize: %w", err)
+	}
+	value, err := normalizeList(req.MergeOutput.Value, req.Source)
+	if err != nil {
+		return contracts.TypedNormalizeResult[dnd.SceneDescriptionList]{}, normalizerErrorf("normalize scenes: %w", err)
+	}
+	return contracts.TypedNormalizeResult[dnd.SceneDescriptionList]{Value: value}, nil
+}
+
+func normalizeList(input dnd.SceneDescriptionList, doc *source.SourceDocument) (dnd.SceneDescriptionList, error) {
+	if doc == nil {
+		return dnd.SceneDescriptionList{}, fmt.Errorf("source document must not be nil")
+	}
+	if input.Scenes == nil {
+		return dnd.SceneDescriptionList{}, fmt.Errorf("scenes must be present")
+	}
+	if len(input.Scenes) == 0 {
+		return dnd.SceneDescriptionList{}, fmt.Errorf("scenes must not be empty")
+	}
+
+	unitPositions := make(map[int]int, len(doc.Units))
+	for index, unit := range doc.Units {
+		unitPositions[unit.ID] = index
+	}
+	output := dnd.SceneDescriptionList{Scenes: make([]dnd.SceneDescription, len(input.Scenes))}
+	for index, scene := range input.Scenes {
+		scene.Title = strings.TrimSpace(scene.Title)
+		scene.Summary = strings.TrimSpace(scene.Summary)
+		if err := shape.Validate(dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{scene}}); err != nil {
+			return dnd.SceneDescriptionList{}, fmt.Errorf("scenes[%d]: %w", index, err)
+		}
+		if err := source.ValidateRef(doc, scene.SourceRef); err != nil {
+			return dnd.SceneDescriptionList{}, fmt.Errorf("scenes[%d].source_ref: %s", index, diagnostics.Truncate(err.Error()))
+		}
+		output.Scenes[index] = scene
+	}
+
+	sort.SliceStable(output.Scenes, func(left, right int) bool {
+		leftStart := unitPositions[output.Scenes[left].SourceRef.StartUnitID]
+		rightStart := unitPositions[output.Scenes[right].SourceRef.StartUnitID]
+		if leftStart != rightStart {
+			return leftStart < rightStart
+		}
+		return output.Scenes[left].ID < output.Scenes[right].ID
+	})
+
+	unique := make([]dnd.SceneDescription, 0, len(output.Scenes))
+	byID := make(map[string]dnd.SceneDescription, len(output.Scenes))
+	byRange := make(map[source.SourceRef]dnd.SceneDescription, len(output.Scenes))
+	seen := make(map[dnd.SceneDescription]struct{}, len(output.Scenes))
+	for _, scene := range output.Scenes {
+		if previous, ok := byID[scene.ID]; ok && !identical(previous, scene) {
+			return dnd.SceneDescriptionList{}, fmt.Errorf("scene ID %s has conflicting records", diagnostics.Quote(scene.ID))
+		}
+		if previous, ok := byRange[scene.SourceRef]; ok && !sameModelContent(previous, scene) {
+			return dnd.SceneDescriptionList{}, fmt.Errorf("source range %s has conflicting records", sourceRefLabel(scene.SourceRef))
+		}
+		if _, ok := seen[scene]; ok {
+			continue
+		}
+		byID[scene.ID] = scene
+		byRange[scene.SourceRef] = scene
+		seen[scene] = struct{}{}
+		unique = append(unique, scene)
+	}
+	output.Scenes = unique
+	return output, nil
+}
+
+func identical(left, right dnd.SceneDescription) bool {
+	return left == right
+}
+
+func sameModelContent(left, right dnd.SceneDescription) bool {
+	return left.Kind == right.Kind && left.Title == right.Title && left.Summary == right.Summary
+}
+
+func sourceRefLabel(ref source.SourceRef) string {
+	return fmt.Sprintf("%s:%d-%d", diagnostics.Quote(ref.SourceID), ref.StartUnitID, ref.EndUnitID)
+}
+
+func ModuleSpec() pipeline.ModuleSpec {
+	return pipeline.ModuleSpec{
+		Key:          Key,
+		Stage:        pipeline.StageNormalize,
+		Requires:     append([]string(nil), requiredCapabilities...),
+		Provides:     append([]string(nil), providedCapabilities...),
+		ArtifactKind: dnd.SceneDescriptionListKind,
+	}
+}
+
+func Register(registry *pipeline.NormalizerRegistry) error {
+	return pipeline.RegisterNormalizerBuilder(registry, ModuleSpec(), validateOptions, func(request pipeline.BuildRequest) (contracts.Normalizer[dnd.SceneDescriptionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(options), nil
+	})
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, normalizerErrorf("%w", err)
+	}
+	return Options{}, nil
+}
+
+func validateOptions(options map[string]any) error { _, err := DecodeOptions(options); return err }
+
+func normalizerErrorf(format string, args ...any) error {
+	return fmt.Errorf("dnd scene descriptions normalizer: "+format, args...)
+}

internal/modules/dnd/normalize/scenedescriptions/normalizer_test.go

@@ -0,0 +1,135 @@
+package scenedescriptions
+
+import (
+	"context"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+)
+
+func TestNormalizeTrimsOrdersDeduplicatesAndOwnsOutput(t *testing.T) {
+	doc := testDocument()
+	input := dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{
+		scene("later", 3, 3, dnd.SceneKindNarrative, "  Return ", " The party returns. "),
+		scene("first-b", 1, 1, dnd.SceneKindMeta, " Rules ", " The table checks rules. "),
+		scene("first-a", 1, 1, dnd.SceneKindMeta, " Rules ", " The table checks rules. "),
+		scene("later", 3, 3, dnd.SceneKindNarrative, "  Return ", " The party returns. "),
+	}}
+	before := cloneList(input)
+	result, err := New(Options{}).Normalize(context.Background(), normalizeRequest(input, doc))
+	if err != nil {
+		t.Fatalf("Normalize() error = %v, want nil", err)
+	}
+	want := []dnd.SceneDescription{
+		scene("first-a", 1, 1, dnd.SceneKindMeta, "Rules", "The table checks rules."),
+		scene("first-b", 1, 1, dnd.SceneKindMeta, "Rules", "The table checks rules."),
+		scene("later", 3, 3, dnd.SceneKindNarrative, "Return", "The party returns."),
+	}
+	if !reflect.DeepEqual(result.Value.Scenes, want) {
+		t.Fatalf("scenes = %#v, want %#v", result.Value.Scenes, want)
+	}
+	if !reflect.DeepEqual(input, before) {
+		t.Fatalf("Normalize() mutated input: %#v", input)
+	}
+	result.Value.Scenes[0].Title = "changed"
+	if input.Scenes[0].Title == "changed" {
+		t.Fatal("normalized output aliases input storage")
+	}
+}
+
+func TestNormalizeRejectsInvalidCandidatesAndConflicts(t *testing.T) {
+	doc := testDocument()
+	valid := scene("one", 1, 1, dnd.SceneKindNarrative, "Arrival", "The party arrives.")
+	for _, test := range []struct {
+		name  string
+		input dnd.SceneDescriptionList
+	}{
+		{name: "nil list", input: dnd.SceneDescriptionList{}},
+		{name: "empty list", input: dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{}}},
+		{name: "blank ID", input: dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{Kind: dnd.SceneKindNarrative, Title: "Arrival", Summary: "The party arrives.", SourceRef: valid.SourceRef}}}},
+		{name: "invalid kind", input: dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{ID: "one", Kind: "other", Title: "Arrival", Summary: "The party arrives.", SourceRef: valid.SourceRef}}}},
+		{name: "invalid range", input: dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{{ID: "one", Kind: dnd.SceneKindNarrative, Title: "Arrival", Summary: "The party arrives.", SourceRef: source.SourceRef{SourceID: doc.ID, StartUnitID: 9, EndUnitID: 9}}}}},
+		{name: "same ID conflict", input: dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{valid, scene("one", 2, 2, dnd.SceneKindNarrative, "Departure", "The party leaves.")}}},
+		{name: "same range conflict", input: dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{valid, scene("two", 1, 1, dnd.SceneKindCombat, "Ambush", "Bandits strike.")}}},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			if _, err := New(Options{}).Normalize(context.Background(), normalizeRequest(test.input, doc)); err == nil {
+				t.Fatal("Normalize() error = nil, want rejection")
+			}
+		})
+	}
+}
+
+func TestNormalizeUsesDocumentOrderRatherThanNumericUnitID(t *testing.T) {
+	doc := &source.SourceDocument{ID: "session", Units: []source.SourceUnit{
+		{ID: 20, Text: "The party arrives."},
+		{ID: 10, Text: "The party departs."},
+	}}
+	input := dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{
+		scene("later", 10, 10, dnd.SceneKindNarrative, "Departure", "The party departs."),
+		scene("first", 20, 20, dnd.SceneKindNarrative, "Arrival", "The party arrives."),
+	}}
+	result, err := New(Options{}).Normalize(context.Background(), normalizeRequest(input, doc))
+	if err != nil {
+		t.Fatalf("Normalize() error = %v", err)
+	}
+	if got, want := []string{result.Value.Scenes[0].ID, result.Value.Scenes[1].ID}, []string{"first", "later"}; !reflect.DeepEqual(got, want) {
+		t.Fatalf("scene order = %#v, want %#v", got, want)
+	}
+}
+
+func TestNormalizerContractAndCancellation(t *testing.T) {
+	if _, err := DecodeOptions(nil); err != nil {
+		t.Fatalf("DecodeOptions(nil) error = %v", err)
+	}
+	if _, err := DecodeOptions(map[string]any{"unexpected": true}); err == nil {
+		t.Fatal("DecodeOptions() accepted unknown options")
+	}
+	want := pipeline.ModuleSpec{Key: Key, Stage: pipeline.StageNormalize, Requires: []string{"merged"}, Provides: []string{"normalized"}, ArtifactKind: dnd.SceneDescriptionListKind}
+	if got := ModuleSpec(); !reflect.DeepEqual(got, want) {
+		t.Fatalf("ModuleSpec() = %#v, want %#v", got, want)
+	}
+	registry := pipeline.NewNormalizerRegistry()
+	if err := Register(registry); err != nil {
+		t.Fatalf("Register() error = %v", err)
+	}
+	normalizer := New(Options{})
+	if metadata := normalizer.ManifestMetadata(); metadata["normalizer_policy"] != normalizerPolicy {
+		t.Fatalf("metadata = %#v", metadata)
+	}
+	if got, want := normalizer.CheckpointFingerprints(), []pipeline.CheckpointFingerprint{{Name: "normalizer_policy", Value: normalizerPolicy}}; !reflect.DeepEqual(got, want) {
+		t.Fatalf("fingerprints = %#v, want %#v", got, want)
+	}
+	canceled, cancel := context.WithCancel(context.Background())
+	cancel()
+	if _, err := normalizer.Normalize(canceled, normalizeRequest(dnd.SceneDescriptionList{}, testDocument())); err == nil || !strings.Contains(err.Error(), "context") {
+		t.Fatalf("canceled Normalize() error = %v", err)
+	}
+	if _, err := normalizer.Normalize(context.Background(), normalizeRequest(dnd.SceneDescriptionList{Scenes: []dnd.SceneDescription{scene("one", 1, 1, dnd.SceneKindNarrative, "Arrival", "The party arrives.")}}, nil)); err == nil || !strings.Contains(err.Error(), "source document") {
+		t.Fatalf("nil source Normalize() error = %v", err)
+	}
+}
+
+func normalizeRequest(value dnd.SceneDescriptionList, doc *source.SourceDocument) contracts.TypedNormalizeRequest[dnd.SceneDescriptionList] {
+	return contracts.TypedNormalizeRequest[dnd.SceneDescriptionList]{Source: doc, MergeOutput: contracts.MergeArtifact[dnd.SceneDescriptionList]{Value: value}}
+}
+
+func testDocument() *source.SourceDocument {
+	return &source.SourceDocument{ID: "session", Units: []source.SourceUnit{{ID: 1, Text: "Rules are discussed."}, {ID: 2, Text: "The party travels."}, {ID: 3, Text: "The party returns."}}}
+}
+
+func scene(id string, start, end int, kind dnd.SceneKind, title, summary string) dnd.SceneDescription {
+	return dnd.SceneDescription{ID: id, SourceRef: source.SourceRef{SourceID: "session", StartUnitID: start, EndUnitID: end}, Kind: kind, Title: title, Summary: summary}
+}
+
+func cloneList(value dnd.SceneDescriptionList) dnd.SceneDescriptionList {
+	if value.Scenes != nil {
+		value.Scenes = append([]dnd.SceneDescription(nil), value.Scenes...)
+	}
+	return value
+}

internal/modules/dnd/npcinteractions/canonical.go

@@ -0,0 +1,166 @@
+// Package npcinteractions owns canonical ordering and exact-identity rules for
+// D&D NPC interaction artifacts.
+package npcinteractions
+
+import (
+	"sort"
+	"strconv"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
+)
+
+// CanonicalizeSourceRefs returns a cloned, document-ordered, de-duplicated
+// source-reference list.
+func CanonicalizeSourceRefs(doc *source.SourceDocument, input []source.SourceRef) []source.SourceRef {
+	if input == nil {
+		return nil
+	}
+	canonical := append([]source.SourceRef(nil), input...)
+	sort.SliceStable(canonical, func(left, right int) bool {
+		return SourceRefLess(doc, canonical[left], canonical[right])
+	})
+	unique := make([]source.SourceRef, 0, len(canonical))
+	for _, ref := range canonical {
+		if len(unique) == 0 || unique[len(unique)-1] != ref {
+			unique = append(unique, ref)
+		}
+	}
+	return unique
+}
+
+// SourceRefsEqual reports whether two source-reference lists have identical
+// representations and values.
+func SourceRefsEqual(left, right []source.SourceRef) bool {
+	if (left == nil) != (right == nil) || len(left) != len(right) {
+		return false
+	}
+	for index := range left {
+		if left[index] != right[index] {
+			return false
+		}
+	}
+	return true
+}
+
+// SourceRefLess orders references by source identity and then by the source
+// document positions of their endpoints. Invalid endpoints sort after valid
+// endpoints and fall back to their literal IDs for deterministic diagnostics.
+func SourceRefLess(doc *source.SourceDocument, left, right source.SourceRef) bool {
+	if left.SourceID != right.SourceID {
+		return left.SourceID < right.SourceID
+	}
+	leftStart, leftStartOK := source.UnitIndex(doc, left.StartUnitID)
+	rightStart, rightStartOK := source.UnitIndex(doc, right.StartUnitID)
+	if leftStartOK != rightStartOK {
+		return leftStartOK
+	}
+	if leftStartOK && leftStart != rightStart {
+		return leftStart < rightStart
+	}
+	if left.StartUnitID != right.StartUnitID {
+		return left.StartUnitID < right.StartUnitID
+	}
+	leftEnd, leftEndOK := source.UnitIndex(doc, left.EndUnitID)
+	rightEnd, rightEndOK := source.UnitIndex(doc, right.EndUnitID)
+	if leftEndOK != rightEndOK {
+		return leftEndOK
+	}
+	if leftEndOK && leftEnd != rightEnd {
+		return leftEnd < rightEnd
+	}
+	return left.EndUnitID < right.EndUnitID
+}
+
+// Less defines the canonical order for NPC interaction occurrences.
+func Less(doc *source.SourceDocument, left, right dnd.NPCInteraction) bool {
+	leftPosition, leftHasEvidence := EarliestSourcePosition(doc, left)
+	rightPosition, rightHasEvidence := EarliestSourcePosition(doc, right)
+	if leftHasEvidence != rightHasEvidence {
+		return leftHasEvidence
+	}
+	if leftHasEvidence && leftPosition != rightPosition {
+		return leftPosition < rightPosition
+	}
+	leftKey := identity.ComparisonKey(left.Name)
+	rightKey := identity.ComparisonKey(right.Name)
+	if leftKey != rightKey {
+		return leftKey < rightKey
+	}
+	if left.Name != right.Name {
+		return left.Name < right.Name
+	}
+	if left.Kind != right.Kind {
+		return left.Kind < right.Kind
+	}
+	return sourceRefsLess(doc, left.SourceRefs, right.SourceRefs)
+}
+
+// EarliestSourcePosition returns the earliest valid cited position.
+func EarliestSourcePosition(doc *source.SourceDocument, interaction dnd.NPCInteraction) (int, bool) {
+	found := false
+	earliest := 0
+	for _, ref := range interaction.SourceRefs {
+		if source.ValidateRef(doc, ref) != nil {
+			continue
+		}
+		position, ok := source.UnitIndex(doc, ref.StartUnitID)
+		if !ok || (found && position >= earliest) {
+			continue
+		}
+		earliest = position
+		found = true
+	}
+	return earliest, found
+}
+
+// ValidSourceRefs reports whether an interaction has non-empty, valid
+// current-document evidence.
+func ValidSourceRefs(doc *source.SourceDocument, refs []source.SourceRef) bool {
+	if len(refs) == 0 {
+		return false
+	}
+	for _, ref := range refs {
+		if source.ValidateRef(doc, ref) != nil {
+			return false
+		}
+	}
+	return true
+}
+
+// ExactIdentity returns a collision-safe key over every durable interaction
+// field. Callers decide whether the record is eligible for duplicate handling.
+func ExactIdentity(interaction dnd.NPCInteraction) string {
+	var key strings.Builder
+	writeKeyString(&key, interaction.Name)
+	writeKeyString(&key, string(interaction.Kind))
+	for _, ref := range interaction.SourceRefs {
+		writeKeyString(&key, ref.SourceID)
+		writeKeyInt(&key, ref.StartUnitID)
+		writeKeyInt(&key, ref.EndUnitID)
+	}
+	return key.String()
+}
+
+func sourceRefsLess(doc *source.SourceDocument, left, right []source.SourceRef) bool {
+	for index := 0; index < len(left) && index < len(right); index++ {
+		if left[index] == right[index] {
+			continue
+		}
+		return SourceRefLess(doc, left[index], right[index])
+	}
+	return len(left) < len(right)
+}
+
+func writeKeyString(builder *strings.Builder, value string) {
+	builder.WriteString(strconv.Itoa(len(value)))
+	builder.WriteByte(':')
+	builder.WriteString(value)
+}
+
+func writeKeyInt(builder *strings.Builder, value int) {
+	builder.WriteString(strconv.Itoa(value))
+	builder.WriteByte(';')
+}

internal/modules/dnd/register/chains.go

@@ -3,19 +3,32 @@ package register
 import (
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
 	combatextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/combatturns"
+	interactionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcinteractions"
 	npcextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcs"
+	scenedescriptionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/scenedescriptions"
 	spellextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/spells"
 	combatnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/combatturns"
+	interactionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcinteractions"
 	npcnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcs"
+	scenedescriptionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/scenedescriptions"
 	spellnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/spells"
 	combatinvariants "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/invariants"
 	combatshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/shape"
 	combatsourcerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/source_refs"
 	combatrelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/source_relatedness"
+	interactioninvariants "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/invariants"
+	interactionregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/registry"
+	interactionshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/shape"
+	interactionrefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/source_refs"
+	interactionrelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/source_relatedness"
 	npcidentity "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/identity"
 	npcshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/shape"
 	npcsourcerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/source_refs"
 	npcrelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/source_relatedness"
+	sceneinvariants "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/invariants"
+	sceneshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/shape"
+	scenerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/source_refs"
+	scenerelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/source_relatedness"
 	spellcatalog "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/spells/catalog"
 	spellshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/spells/shape"
 	spellsourcerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/spells/source_refs"
@@ -108,5 +121,61 @@ func registerDefaultChains(registry *pipeline.ValidatorChainRegistry) error {
 				},
 			})
 		}},
+		{name: "npc interactions validator chain", register: func() error {
+			return registry.Register(pipeline.ValidatorChainMapping{
+				Stage:  pipeline.StageExtract,
+				Module: interactionextract.Key,
+				Validators: []pipeline.ModuleBinding{
+					pipeline.Binding(validjson.Key),
+					pipeline.Binding(interactionshape.Key),
+					pipeline.Binding(interactionregistry.Key),
+					pipeline.Binding(interactionrefs.Key),
+					pipeline.Binding(validjsonschema.Key),
+					pipeline.Binding(interactionrelatedness.Key),
+				},
+			})
+		}},
+		{name: "npc interactions normalize validator chain", register: func() error {
+			return registry.Register(pipeline.ValidatorChainMapping{
+				Stage:  pipeline.StageNormalize,
+				Module: interactionnormalize.Key,
+				Validators: []pipeline.ModuleBinding{
+					pipeline.Binding(validjson.Key),
+					pipeline.Binding(interactionshape.Key),
+					pipeline.Binding(interactionregistry.Key),
+					pipeline.Binding(interactioninvariants.Key),
+					pipeline.Binding(interactionrefs.Key),
+					pipeline.Binding(validjsonschema.Key),
+					pipeline.Binding(interactionrelatedness.Key),
+				},
+			})
+		}},
+		{name: "scene descriptions validator chain", register: func() error {
+			return registry.Register(pipeline.ValidatorChainMapping{
+				Stage:  pipeline.StageExtract,
+				Module: scenedescriptionextract.Key,
+				Validators: []pipeline.ModuleBinding{
+					pipeline.Binding(validjson.Key),
+					pipeline.Binding(sceneshape.Key),
+					pipeline.Binding(scenerefs.Key),
+					pipeline.Binding(validjsonschema.Key),
+					pipeline.Binding(scenerelatedness.Key),
+				},
+			})
+		}},
+		{name: "scene descriptions normalize validator chain", register: func() error {
+			return registry.Register(pipeline.ValidatorChainMapping{
+				Stage:  pipeline.StageNormalize,
+				Module: scenedescriptionnormalize.Key,
+				Validators: []pipeline.ModuleBinding{
+					pipeline.Binding(validjson.Key),
+					pipeline.Binding(sceneshape.Key),
+					pipeline.Binding(sceneinvariants.Key),
+					pipeline.Binding(scenerefs.Key),
+					pipeline.Binding(validjsonschema.Key),
+					pipeline.Binding(scenerelatedness.Key),
+				},
+			})
+		}},
 	})
 }

internal/modules/dnd/register/merge.go

@@ -57,6 +57,46 @@ func appendCombatTurnLists(values []dnd.CombatTurnList) (dnd.CombatTurnList, err
 	return combined, nil
 }
 
+func appendNPCInteractionLists(values []dnd.NPCInteractionList) (dnd.NPCInteractionList, error) {
+	count := 0
+	present := false
+	for _, value := range values {
+		if value.Interactions != nil {
+			present = true
+		}
+		count += len(value.Interactions)
+	}
+	if !present {
+		return dnd.NPCInteractionList{}, nil
+	}
+	combined := dnd.NPCInteractionList{Interactions: make([]dnd.NPCInteraction, 0, count)}
+	for _, value := range values {
+		for _, interaction := range value.Interactions {
+			combined.Interactions = append(combined.Interactions, cloneNPCInteraction(interaction))
+		}
+	}
+	return combined, nil
+}
+
+func appendSceneDescriptionLists(values []dnd.SceneDescriptionList) (dnd.SceneDescriptionList, error) {
+	count := 0
+	present := false
+	for _, value := range values {
+		if value.Scenes != nil {
+			present = true
+		}
+		count += len(value.Scenes)
+	}
+	if !present {
+		return dnd.SceneDescriptionList{}, nil
+	}
+	combined := dnd.SceneDescriptionList{Scenes: make([]dnd.SceneDescription, 0, count)}
+	for _, value := range values {
+		combined.Scenes = append(combined.Scenes, value.Scenes...)
+	}
+	return combined, nil
+}
+
 func cloneCombatTurn(value dnd.CombatTurn) dnd.CombatTurn {
 	clone := value
 	if value.SourceRefs != nil {
@@ -64,3 +104,11 @@ func cloneCombatTurn(value dnd.CombatTurn) dnd.CombatTurn {
 	}
 	return clone
 }
+
+func cloneNPCInteraction(value dnd.NPCInteraction) dnd.NPCInteraction {
+	clone := value
+	if value.SourceRefs != nil {
+		clone.SourceRefs = append([]source.SourceRef(nil), value.SourceRefs...)
+	}
+	return clone
+}

internal/modules/dnd/register/modules.go

@@ -6,13 +6,19 @@ import (
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/chunk/scenes"
 	combatcodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/combatturns"
+	interactioncodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcinteractions"
 	npccodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcs"
+	scenedescriptioncodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/scenedescriptions"
 	spellcodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/spells"
 	combatextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/combatturns"
+	interactionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcinteractions"
 	npcextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcs"
+	scenedescriptionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/scenedescriptions"
 	spellextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/spells"
 	combatnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/combatturns"
+	interactionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcinteractions"
 	npcnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcs"
+	scenedescriptionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/scenedescriptions"
 	spellnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/spells"
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/generic/merge/appendorder"
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/generic/normalize/noop"
@@ -24,10 +30,16 @@ func registerModules(registries pipeline.Registries) error {
 		{name: "spells codec", register: func() error { return pipeline.RegisterArtifactCodec(registries.ArtifactCodecs, codec) }},
 		{name: "npcs codec", register: func() error { return pipeline.RegisterArtifactCodec(registries.ArtifactCodecs, npccodec.New()) }},
 		{name: "combat turns codec", register: func() error { return pipeline.RegisterArtifactCodec(registries.ArtifactCodecs, combatcodec.New()) }},
+		{name: "npc interactions codec", register: func() error { return pipeline.RegisterArtifactCodec(registries.ArtifactCodecs, interactioncodec.New()) }},
+		{name: "scene descriptions codec", register: func() error {
+			return pipeline.RegisterArtifactCodec(registries.ArtifactCodecs, scenedescriptioncodec.New())
+		}},
 		{name: "scenes chunker", register: func() error { return scenes.Register(registries.Chunkers) }},
 		{name: "spells extractor", register: func() error { return spellextract.Register(registries.Extractors) }},
 		{name: "npcs extractor", register: func() error { return npcextract.Register(registries.Extractors) }},
 		{name: "combat turns extractor", register: func() error { return combatextract.Register(registries.Extractors) }},
+		{name: "npc interactions extractor", register: func() error { return interactionextract.Register(registries.Extractors) }},
+		{name: "scene descriptions extractor", register: func() error { return scenedescriptionextract.Register(registries.Extractors) }},
 		{name: "spell-list appendorder merger", register: func() error {
 			return appendorder.RegisterTyped(registries.Mergers, dnd.SpellListKind, appendSpellLists)
 		}},
@@ -37,9 +49,17 @@ func registerModules(registries pipeline.Registries) error {
 		{name: "combat-turn-list appendorder merger", register: func() error {
 			return appendorder.RegisterTyped(registries.Mergers, dnd.CombatTurnListKind, appendCombatTurnLists)
 		}},
+		{name: "npc-interaction-list appendorder merger", register: func() error {
+			return appendorder.RegisterTyped(registries.Mergers, dnd.NPCInteractionListKind, appendNPCInteractionLists)
+		}},
+		{name: "scene-description-list appendorder merger", register: func() error {
+			return appendorder.RegisterTyped(registries.Mergers, dnd.SceneDescriptionListKind, appendSceneDescriptionLists)
+		}},
 		{name: "spells normalizer", register: func() error { return spellnormalize.Register(registries.Normalizers) }},
 		{name: "npcs normalizer", register: func() error { return npcnormalize.Register(registries.Normalizers) }},
 		{name: "combat turns normalizer", register: func() error { return combatnormalize.Register(registries.Normalizers) }},
+		{name: "npc interactions normalizer", register: func() error { return interactionnormalize.Register(registries.Normalizers) }},
+		{name: "scene descriptions normalizer", register: func() error { return scenedescriptionnormalize.Register(registries.Normalizers) }},
 		{name: "spell-list noop normalizer", register: func() error {
 			return noop.RegisterTyped[dnd.SpellList](registries.Normalizers, dnd.SpellListKind)
 		}},
@@ -49,6 +69,12 @@ func registerModules(registries pipeline.Registries) error {
 		{name: "combat-turn-list noop normalizer", register: func() error {
 			return noop.RegisterTyped[dnd.CombatTurnList](registries.Normalizers, dnd.CombatTurnListKind)
 		}},
+		{name: "npc-interaction-list noop normalizer", register: func() error {
+			return noop.RegisterTyped[dnd.NPCInteractionList](registries.Normalizers, dnd.NPCInteractionListKind)
+		}},
+		{name: "scene-description-list noop normalizer", register: func() error {
+			return noop.RegisterTyped[dnd.SceneDescriptionList](registries.Normalizers, dnd.SceneDescriptionListKind)
+		}},
 	})
 }
 
@@ -58,5 +84,7 @@ func registerPromptAssets(assets *llm.AssetRegistry) error {
 		{name: "spells prompt assets", register: func() error { return spellextract.RegisterPromptAssets(assets) }},
 		{name: "npcs prompt assets", register: func() error { return npcextract.RegisterPromptAssets(assets) }},
 		{name: "combat turns prompt assets", register: func() error { return combatextract.RegisterPromptAssets(assets) }},
+		{name: "npc interactions prompt assets", register: func() error { return interactionextract.RegisterPromptAssets(assets) }},
+		{name: "scene descriptions prompt assets", register: func() error { return scenedescriptionextract.RegisterPromptAssets(assets) }},
 	})
 }

internal/modules/dnd/register/register_test.go

@@ -13,10 +13,14 @@ import (
 	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
 	combatextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/combatturns"
+	interactionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcinteractions"
 	npcextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/npcs"
+	scenedescriptionextract "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/scenedescriptions"
 	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/extract/spells"
 	combatnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/combatturns"
+	interactionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcinteractions"
 	npcnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/npcs"
+	scenedescriptionnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/scenedescriptions"
 	spellnormalize "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/normalize/spells"
 )
 
@@ -27,13 +31,15 @@ func TestRegisterAddsDNDFamily(t *testing.T) {
 		t.Fatalf("Register() error = %v, want nil", err)
 	}
 	assertContainsKeys(t, "chunkers", registries.Chunkers.RegisteredKeys(), []string{"dnd/scenes"})
-	assertContainsKeys(t, "extractors", registries.Extractors.RegisteredKeys(), []string{"dnd/spells", npcextract.Key, combatextract.Key})
-	assertContainsKeys(t, "normalizers", registries.Normalizers.RegisteredKeys(), []string{spellnormalize.Key, npcnormalize.Key, combatnormalize.Key, pipeline.DefaultNormalizeModule})
-	assertContainsArtifactKinds(t, registries.ArtifactCodecs.RegisteredKinds(), []contracts.ArtifactKind{dnd.SpellListKind, dnd.NPCListKind, dnd.CombatTurnListKind})
-	assertContainsArtifactKinds(t, registries.Mergers.RegisteredArtifactKinds(pipeline.DefaultMergeModule), []contracts.ArtifactKind{dnd.SpellListKind, dnd.NPCListKind, dnd.CombatTurnListKind})
-	assertContainsArtifactKinds(t, registries.Normalizers.RegisteredArtifactKinds(pipeline.DefaultNormalizeModule), []contracts.ArtifactKind{dnd.SpellListKind, dnd.NPCListKind, dnd.CombatTurnListKind})
+	assertContainsKeys(t, "extractors", registries.Extractors.RegisteredKeys(), []string{"dnd/spells", npcextract.Key, combatextract.Key, interactionextract.Key, scenedescriptionextract.Key})
+	assertContainsKeys(t, "normalizers", registries.Normalizers.RegisteredKeys(), []string{spellnormalize.Key, npcnormalize.Key, combatnormalize.Key, interactionnormalize.Key, scenedescriptionnormalize.Key, pipeline.DefaultNormalizeModule})
+	assertContainsArtifactKinds(t, registries.ArtifactCodecs.RegisteredKinds(), []contracts.ArtifactKind{dnd.SpellListKind, dnd.NPCListKind, dnd.CombatTurnListKind, dnd.NPCInteractionListKind, dnd.SceneDescriptionListKind})
+	assertContainsArtifactKinds(t, registries.Mergers.RegisteredArtifactKinds(pipeline.DefaultMergeModule), []contracts.ArtifactKind{dnd.SpellListKind, dnd.NPCListKind, dnd.CombatTurnListKind, dnd.NPCInteractionListKind, dnd.SceneDescriptionListKind})
+	assertContainsArtifactKinds(t, registries.Normalizers.RegisteredArtifactKinds(pipeline.DefaultNormalizeModule), []contracts.ArtifactKind{dnd.SpellListKind, dnd.NPCListKind, dnd.CombatTurnListKind, dnd.NPCInteractionListKind, dnd.SceneDescriptionListKind})
 	assertContainsArtifactKinds(t, registries.Normalizers.RegisteredArtifactKinds(npcnormalize.Key), []contracts.ArtifactKind{dnd.NPCListKind})
 	assertContainsArtifactKinds(t, registries.Normalizers.RegisteredArtifactKinds(combatnormalize.Key), []contracts.ArtifactKind{dnd.CombatTurnListKind})
+	assertContainsArtifactKinds(t, registries.Normalizers.RegisteredArtifactKinds(interactionnormalize.Key), []contracts.ArtifactKind{dnd.NPCInteractionListKind})
+	assertContainsArtifactKinds(t, registries.Normalizers.RegisteredArtifactKinds(scenedescriptionnormalize.Key), []contracts.ArtifactKind{dnd.SceneDescriptionListKind})
 	assertContainsKeys(t, "validators", registries.Validators.RegisteredKeys(), []string{
 		"extract/dnd/npcs/shape",
 		"extract/dnd/npcs/source_refs",
@@ -47,6 +53,15 @@ func TestRegisterAddsDNDFamily(t *testing.T) {
 		"extract/dnd/combat-turns/source_refs",
 		"extract/dnd/combat-turns/source_relatedness",
 		"normalize/dnd/combat-turns/invariants",
+		"extract/dnd/npc-interactions/shape",
+		"extract/dnd/npc-interactions/registry",
+		"extract/dnd/npc-interactions/source_refs",
+		"extract/dnd/npc-interactions/source_relatedness",
+		"normalize/dnd/npc-interactions/invariants",
+		"extract/dnd/scene-descriptions/shape",
+		"extract/dnd/scene-descriptions/source_refs",
+		"extract/dnd/scene-descriptions/source_relatedness",
+		"normalize/dnd/scene-descriptions/invariants",
 		"generic/always_accept",
 		"generic/always_reject",
 	})
@@ -106,9 +121,56 @@ func TestRegisterAddsDNDFamily(t *testing.T) {
 	if got := registries.ValidatorChains.Validators(pipeline.StageNormalize, combatnormalize.Key); !reflect.DeepEqual(got, combatNormalizeChain) {
 		t.Fatalf("combat normalize validator chain = %#v, want %#v", got, combatNormalizeChain)
 	}
+	interactionExtractChain := []pipeline.ModuleBinding{
+		pipeline.Binding("generic/valid_json"),
+		pipeline.Binding("extract/dnd/npc-interactions/shape"),
+		pipeline.Binding("extract/dnd/npc-interactions/registry"),
+		pipeline.Binding("extract/dnd/npc-interactions/source_refs"),
+		pipeline.Binding("generic/valid_json_schema"),
+		pipeline.Binding("extract/dnd/npc-interactions/source_relatedness"),
+	}
+	interactionNormalizeChain := []pipeline.ModuleBinding{
+		pipeline.Binding("generic/valid_json"),
+		pipeline.Binding("extract/dnd/npc-interactions/shape"),
+		pipeline.Binding("extract/dnd/npc-interactions/registry"),
+		pipeline.Binding("normalize/dnd/npc-interactions/invariants"),
+		pipeline.Binding("extract/dnd/npc-interactions/source_refs"),
+		pipeline.Binding("generic/valid_json_schema"),
+		pipeline.Binding("extract/dnd/npc-interactions/source_relatedness"),
+	}
+	if got := registries.ValidatorChains.Validators(pipeline.StageExtract, interactionextract.Key); !reflect.DeepEqual(got, interactionExtractChain) {
+		t.Fatalf("NPC interaction extract validator chain = %#v, want %#v", got, interactionExtractChain)
+	}
+	if got := registries.ValidatorChains.Validators(pipeline.StageNormalize, interactionnormalize.Key); !reflect.DeepEqual(got, interactionNormalizeChain) {
+		t.Fatalf("NPC interaction normalize validator chain = %#v, want %#v", got, interactionNormalizeChain)
+	}
+	sceneExtractChain := []pipeline.ModuleBinding{
+		pipeline.Binding("generic/valid_json"),
+		pipeline.Binding("extract/dnd/scene-descriptions/shape"),
+		pipeline.Binding("extract/dnd/scene-descriptions/source_refs"),
+		pipeline.Binding("generic/valid_json_schema"),
+		pipeline.Binding("extract/dnd/scene-descriptions/source_relatedness"),
+	}
+	if got := registries.ValidatorChains.Validators(pipeline.StageExtract, scenedescriptionextract.Key); !reflect.DeepEqual(got, sceneExtractChain) {
+		t.Fatalf("scene description extract validator chain = %#v, want %#v", got, sceneExtractChain)
+	}
+	sceneNormalizeChain := []pipeline.ModuleBinding{
+		pipeline.Binding("generic/valid_json"),
+		pipeline.Binding("extract/dnd/scene-descriptions/shape"),
+		pipeline.Binding("normalize/dnd/scene-descriptions/invariants"),
+		pipeline.Binding("extract/dnd/scene-descriptions/source_refs"),
+		pipeline.Binding("generic/valid_json_schema"),
+		pipeline.Binding("extract/dnd/scene-descriptions/source_relatedness"),
+	}
+	if got := registries.ValidatorChains.Validators(pipeline.StageNormalize, scenedescriptionnormalize.Key); !reflect.DeepEqual(got, sceneNormalizeChain) {
+		t.Fatalf("scene description normalize validator chain = %#v, want %#v", got, sceneNormalizeChain)
+	}
 	if got := registries.ValidatorChains.Validators(pipeline.StageMerge, npcextract.Key); got != nil {
 		t.Fatalf("NPC merge validator chain = %#v, want absent", got)
 	}
+	if got := registries.ValidatorChains.Validators(pipeline.StageMerge, interactionextract.Key); got != nil {
+		t.Fatalf("NPC interaction merge validator chain = %#v, want absent", got)
+	}
 	assertAssetNamesContain(t, assets.PromptFS, []string{
 		"dnd.scenes/dnd.scenes.yaml",
 		"dnd.scenes/instructions.md",
@@ -134,12 +196,25 @@ func TestRegisterAddsDNDFamily(t *testing.T) {
 		"dnd.combat_turns/sharedassets/common-dnd-system.md",
 		"dnd.combat_turns/sharedassets/common-dnd-transcript.md",
 		"dnd.combat_turns/task.md",
+		"dnd.npc_interactions/dnd.npc_interactions.yaml",
+		"dnd.npc_interactions/instructions.md",
+		"dnd.npc_interactions/sharedassets/common-dnd-npcs.md",
+		"dnd.npc_interactions/task.md",
+		"dnd.scene_descriptions/dnd.scene_descriptions.yaml",
+		"dnd.scene_descriptions/instructions.md",
+		"dnd.scene_descriptions/sharedassets/common-dnd-identity.md",
+		"dnd.scene_descriptions/sharedassets/common-dnd-references.md",
+		"dnd.scene_descriptions/sharedassets/common-dnd-system.md",
+		"dnd.scene_descriptions/sharedassets/common-dnd-transcript.md",
+		"dnd.scene_descriptions/task.md",
 	})
 	assertAssetNamesContain(t, assets.SchemaFS, []string{
 		"dnd_scenes.v1.json",
 		"dnd_spells_llm.v1.json",
 		"dnd_npcs_llm.v1.json",
 		"dnd_combat_turns_llm.v1.json",
+		"dnd_npc_interactions_llm.v1.json",
+		"dnd_scene_descriptions_llm.v1.json",
 	})
 	if spec, ok := registries.Chunkers.Spec("dnd/scenes"); !ok || spec.Key != "dnd/scenes" {
 		t.Fatalf("scene chunker spec = %#v, present = %t; want family-owned spec", spec, ok)
@@ -162,6 +237,33 @@ func TestRegisterAddsDNDFamily(t *testing.T) {
 	if spec, ok := registries.Normalizers.Spec(combatnormalize.Key); !ok || spec.ArtifactKind != dnd.CombatTurnListKind || spec.Stage != pipeline.StageNormalize {
 		t.Fatalf("combat normalizer spec = %#v, present = %t; want dnd combat-turn-list artifact", spec, ok)
 	}
+	interactionExtractSpec, extractOK := registries.Extractors.Spec(interactionextract.Key)
+	interactionNormalizeSpec, normalizeOK := registries.Normalizers.Spec(interactionnormalize.Key)
+	if !extractOK || interactionExtractSpec.ArtifactKind != dnd.NPCInteractionListKind || !normalizeOK || interactionNormalizeSpec.ArtifactKind != dnd.NPCInteractionListKind || interactionNormalizeSpec.Stage != pipeline.StageNormalize {
+		t.Fatalf("NPC interaction specs = %#v / %#v, present = %t / %t", interactionExtractSpec, interactionNormalizeSpec, extractOK, normalizeOK)
+	}
+	sceneExtractSpec, sceneExtractOK := registries.Extractors.Spec(scenedescriptionextract.Key)
+	sceneNormalizeSpec, sceneNormalizeOK := registries.Normalizers.Spec(scenedescriptionnormalize.Key)
+	if !sceneExtractOK || sceneExtractSpec.ArtifactKind != dnd.SceneDescriptionListKind || !sceneNormalizeOK || sceneNormalizeSpec.ArtifactKind != dnd.SceneDescriptionListKind || sceneNormalizeSpec.Stage != pipeline.StageNormalize {
+		t.Fatalf("scene description specs = %#v / %#v, present = %t / %t", sceneExtractSpec, sceneNormalizeSpec, sceneExtractOK, sceneNormalizeOK)
+	}
+	if len(sceneExtractSpec.ReferenceSlots) != 3 || len(sceneNormalizeSpec.ReferenceSlots) != 0 {
+		t.Fatalf("scene description reference slots = %#v / %#v, want extractor campaign slots only", sceneExtractSpec.ReferenceSlots, sceneNormalizeSpec.ReferenceSlots)
+	}
+	extractRegistrySlot := referenceSlot(interactionExtractSpec.ReferenceSlots, "npcs")
+	normalizeRegistrySlot := referenceSlot(interactionNormalizeSpec.ReferenceSlots, "npcs")
+	if !extractRegistrySlot.Required || !reflect.DeepEqual(extractRegistrySlot.AcceptedArtifactKinds, []contracts.ArtifactKind{dnd.NPCListKind}) || !reflect.DeepEqual(extractRegistrySlot, normalizeRegistrySlot) {
+		t.Fatalf("NPC interaction registry slots disagree: %#v / %#v", interactionExtractSpec.ReferenceSlots, interactionNormalizeSpec.ReferenceSlots)
+	}
+}
+
+func referenceSlot(slots []contracts.ReferenceSlot, name string) contracts.ReferenceSlot {
+	for _, slot := range slots {
+		if slot.Name == name {
+			return slot
+		}
+	}
+	return contracts.ReferenceSlot{}
 }
 
 func TestAppendNPCListsPreservesOrderAndArrayPresence(t *testing.T) {
@@ -185,6 +287,52 @@ func TestAppendNPCListsPreservesOrderAndArrayPresence(t *testing.T) {
 	}
 }
 
+func TestAppendNPCInteractionListsPreservesOrderPresenceAndOwnership(t *testing.T) {
+	refs := []source.SourceRef{{SourceID: "session", StartUnitID: 10, EndUnitID: 10}}
+	input := []dnd.NPCInteractionList{
+		{},
+		{Interactions: []dnd.NPCInteraction{}},
+		{Interactions: []dnd.NPCInteraction{{Name: "Aria", Kind: dnd.NPCInteractionKindDialogue, SourceRefs: refs}}},
+		{Interactions: []dnd.NPCInteraction{{Name: "Borin", Kind: dnd.NPCInteractionKindMentioned, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 20, EndUnitID: 20}}}}},
+	}
+	got, err := appendNPCInteractionLists(input)
+	if err != nil {
+		t.Fatalf("appendNPCInteractionLists() error = %v", err)
+	}
+	if got.Interactions == nil || !reflect.DeepEqual([]string{got.Interactions[0].Name, got.Interactions[1].Name}, []string{"Aria", "Borin"}) {
+		t.Fatalf("combined interactions = %#v", got)
+	}
+	got.Interactions[0].SourceRefs[0].StartUnitID = 999
+	if input[2].Interactions[0].SourceRefs[0].StartUnitID == 999 {
+		t.Fatal("merged interactions share source reference storage")
+	}
+	for _, values := range [][]dnd.NPCInteractionList{nil, []dnd.NPCInteractionList{{}, {}}} {
+		result, err := appendNPCInteractionLists(values)
+		if err != nil || result.Interactions != nil {
+			t.Fatalf("nil-only merge = %#v, %v; want nil interactions", result, err)
+		}
+	}
+}
+
+func TestAppendSceneDescriptionListsPreservesOrderAndArrayPresence(t *testing.T) {
+	input := []dnd.SceneDescriptionList{
+		{},
+		{Scenes: []dnd.SceneDescription{}},
+		{Scenes: []dnd.SceneDescription{{ID: "first"}}},
+		{Scenes: []dnd.SceneDescription{{ID: "second"}}},
+	}
+	got, err := appendSceneDescriptionLists(input)
+	if err != nil || !reflect.DeepEqual(got.Scenes, []dnd.SceneDescription{{ID: "first"}, {ID: "second"}}) {
+		t.Fatalf("appendSceneDescriptionLists() = %#v, error = %v", got, err)
+	}
+	for _, values := range [][]dnd.SceneDescriptionList{nil, []dnd.SceneDescriptionList{{}, {}}} {
+		result, err := appendSceneDescriptionLists(values)
+		if err != nil || result.Scenes != nil {
+			t.Fatalf("nil-only merge = %#v, %v; want nil scenes", result, err)
+		}
+	}
+}
+
 func TestAppendCombatTurnListsPreservesOrderPresenceAndOwnership(t *testing.T) {
 	refs := []source.SourceRef{{SourceID: "session", StartUnitID: 1, EndUnitID: 1}}
 	input := []dnd.CombatTurnList{

internal/modules/dnd/register/validators.go

@@ -7,10 +7,19 @@ import (
 	combatshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/shape"
 	combatsourcerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/source_refs"
 	combatrelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/combatturns/source_relatedness"
+	interactioninvariants "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/invariants"
+	interactionregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/registry"
+	interactionshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/shape"
+	interactionrefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/source_refs"
+	interactionrelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/source_relatedness"
 	npcidentity "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/identity"
 	npcshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/shape"
 	npcsourcerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/source_refs"
 	npcrelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcs/source_relatedness"
+	sceneinvariants "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/invariants"
+	sceneshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/shape"
+	scenerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/source_refs"
+	scenerelatedness "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/scenedescriptions/source_relatedness"
 	spellcatalog "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/spells/catalog"
 	spellshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/spells/shape"
 	spellsourcerefs "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/spells/source_refs"
@@ -33,6 +42,15 @@ func registerValidators(registries pipeline.Registries) error {
 		{name: "combat source references validator", register: func() error { return combatsourcerefs.Register(registries.Validators) }},
 		{name: "combat source relatedness validator", register: func() error { return combatrelatedness.Register(registries.Validators) }},
 		{name: "combat normalized invariants validator", register: func() error { return combatinvariants.Register(registries.Validators) }},
+		{name: "npc interaction shape validator", register: func() error { return interactionshape.Register(registries.Validators) }},
+		{name: "npc interaction registry validator", register: func() error { return interactionregistry.Register(registries.Validators) }},
+		{name: "npc interaction source references validator", register: func() error { return interactionrefs.Register(registries.Validators) }},
+		{name: "npc interaction source relatedness validator", register: func() error { return interactionrelatedness.Register(registries.Validators) }},
+		{name: "npc interaction normalized invariants validator", register: func() error { return interactioninvariants.Register(registries.Validators) }},
+		{name: "scene description shape validator", register: func() error { return sceneshape.Register(registries.Validators) }},
+		{name: "scene description source references validator", register: func() error { return scenerefs.Register(registries.Validators) }},
+		{name: "scene description source relatedness validator", register: func() error { return scenerelatedness.Register(registries.Validators) }},
+		{name: "scene description normalized invariants validator", register: func() error { return sceneinvariants.Register(registries.Validators) }},
 		{name: "spell-list always accept validator", register: func() error {
 			return alwaysaccept.RegisterTyped[dnd.SpellList](registries.Validators, dnd.SpellListKind)
 		}},
@@ -51,5 +69,17 @@ func registerValidators(registries pipeline.Registries) error {
 		{name: "combat-turn-list always reject validator", register: func() error {
 			return alwaysreject.RegisterTyped[dnd.CombatTurnList](registries.Validators, dnd.CombatTurnListKind)
 		}},
+		{name: "npc-interaction-list always accept validator", register: func() error {
+			return alwaysaccept.RegisterTyped[dnd.NPCInteractionList](registries.Validators, dnd.NPCInteractionListKind)
+		}},
+		{name: "npc-interaction-list always reject validator", register: func() error {
+			return alwaysreject.RegisterTyped[dnd.NPCInteractionList](registries.Validators, dnd.NPCInteractionListKind)
+		}},
+		{name: "scene-description-list always accept validator", register: func() error {
+			return alwaysaccept.RegisterTyped[dnd.SceneDescriptionList](registries.Validators, dnd.SceneDescriptionListKind)
+		}},
+		{name: "scene-description-list always reject validator", register: func() error {
+			return alwaysreject.RegisterTyped[dnd.SceneDescriptionList](registries.Validators, dnd.SceneDescriptionListKind)
+		}},
 	})
 }

internal/modules/dnd/shared/assets/prompts/common-dnd-npcs.md

@@ -1,6 +1,6 @@
-An optional normalized Dungeons & Dragons NPC registry is provided below as
-grounding material. Use it only to prefer exact canonical participant names
-and recognize their aliases when the transcript identifies a participant.
+A normalized Dungeons & Dragons NPC registry is provided below as grounding
+material. It may be empty. Use it only to prefer exact canonical participant
+names when the transcript identifies a participant.
 
 Registry content is context, not event evidence. Do not extract events,
 participants, effects, or source references from the registry. Registry source

internal/modules/dnd/shared/diagnostics/diagnostics.go

@@ -6,10 +6,13 @@ import (
 	"fmt"
 	"strconv"
 	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
 )
 
 const (
 	MaxIssues         = 20
+	MaxWarnings       = 20
 	MaxDisplayedRunes = 128
 	MaxMessageBytes   = 4096
 )
@@ -37,6 +40,28 @@ func Aggregate(prefix string, issues []string) string {
 	return aggregateMessage(prefix, displayed, len(issues)-len(displayed))
 }
 
+// LimitWarnings returns at most MaxWarnings warnings, reserving the final
+// position for a deterministic omission summary when truncation is required.
+func LimitWarnings(warnings []contracts.Warning, scope, reasonCode string) []contracts.Warning {
+	if warnings == nil {
+		return nil
+	}
+	if len(warnings) <= MaxWarnings {
+		bounded := make([]contracts.Warning, len(warnings))
+		copy(bounded, warnings)
+		return bounded
+	}
+	displayed := MaxWarnings - 1
+	bounded := make([]contracts.Warning, displayed, MaxWarnings)
+	copy(bounded, warnings[:displayed])
+	bounded = append(bounded, contracts.Warning{
+		Scope:      scope,
+		ReasonCode: reasonCode,
+		Message:    fmt.Sprintf("%d additional warning(s) omitted", len(warnings)-displayed),
+	})
+	return bounded
+}
+
 func aggregateMessage(prefix string, issues []string, omitted int) string {
 	message := prefix + ": " + strings.Join(issues, ", ")
 	if omitted > 0 {

internal/modules/dnd/shared/diagnostics/diagnostics_test.go

@@ -2,9 +2,12 @@ package diagnostics
 
 import (
 	"fmt"
+	"reflect"
 	"strings"
 	"testing"
 	"unicode/utf8"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
 )
 
 func TestAggregateEnforcesByteBudgetAndReportsOmissions(t *testing.T) {
@@ -26,3 +29,25 @@ func TestAggregateEnforcesByteBudgetAndReportsOmissions(t *testing.T) {
 		t.Fatalf("Aggregate() = %q, want %q", message, wantOmitted)
 	}
 }
+
+func TestLimitWarningsBoundsOutputAndReportsOmissions(t *testing.T) {
+	warnings := make([]contracts.Warning, MaxWarnings+3)
+	for index := range warnings {
+		warnings[index] = contracts.Warning{ReasonCode: fmt.Sprintf("warning-%d", index)}
+	}
+	before := append([]contracts.Warning(nil), warnings...)
+
+	got := LimitWarnings(warnings, "records", "warnings_omitted")
+	if len(got) != MaxWarnings {
+		t.Fatalf("LimitWarnings() count = %d, want %d", len(got), MaxWarnings)
+	}
+	summary := got[len(got)-1]
+	wantOmitted := len(warnings) - (MaxWarnings - 1)
+	if summary.Scope != "records" || summary.ReasonCode != "warnings_omitted" ||
+		summary.Message != fmt.Sprintf("%d additional warning(s) omitted", wantOmitted) {
+		t.Fatalf("summary = %#v", summary)
+	}
+	if !reflect.DeepEqual(warnings, before) {
+		t.Fatal("LimitWarnings() mutated its input")
+	}
+}

internal/modules/dnd/types.go

@@ -12,6 +12,10 @@ const NPCListKind contracts.ArtifactKind = "dnd/npc-list"
 
 const CombatTurnListKind contracts.ArtifactKind = "dnd/combat-turn-list"
 
+const NPCInteractionListKind contracts.ArtifactKind = "dnd/npc-interaction-list"
+
+const SceneDescriptionListKind contracts.ArtifactKind = "dnd/scene-description-list"
+
 type SpellList struct {
 	SpellCasts []SpellCast `json:"spell_casts"`
 }
@@ -51,3 +55,45 @@ type CombatTurn struct {
 	TurnKind   CombatTurnKind     `json:"turn_kind"`
 	SourceRefs []source.SourceRef `json:"source_refs"`
 }
+
+type NPCInteractionKind string
+
+const (
+	NPCInteractionKindMentioned         NPCInteractionKind = "mentioned"
+	NPCInteractionKindNoncombatPresence NPCInteractionKind = "noncombat_presence"
+	NPCInteractionKindDialogue          NPCInteractionKind = "dialogue"
+	NPCInteractionKindCombatAlly        NPCInteractionKind = "combat_ally"
+	NPCInteractionKindCombatOpponent    NPCInteractionKind = "combat_opponent"
+	NPCInteractionKindOther             NPCInteractionKind = "other"
+)
+
+type NPCInteractionList struct {
+	Interactions []NPCInteraction `json:"interactions"`
+}
+
+type NPCInteraction struct {
+	Name       string             `json:"name"`
+	Kind       NPCInteractionKind `json:"kind"`
+	SourceRefs []source.SourceRef `json:"source_refs"`
+}
+
+type SceneKind string
+
+const (
+	SceneKindCombat    SceneKind = "combat"
+	SceneKindNarrative SceneKind = "narrative"
+	SceneKindRecap     SceneKind = "recap"
+	SceneKindMeta      SceneKind = "meta"
+)
+
+type SceneDescriptionList struct {
+	Scenes []SceneDescription `json:"scenes"`
+}
+
+type SceneDescription struct {
+	ID        string           `json:"id"`
+	SourceRef source.SourceRef `json:"source_ref"`
+	Kind      SceneKind        `json:"kind"`
+	Title     string           `json:"title"`
+	Summary   string           `json:"summary"`
+}

internal/modules/dnd/validate/npcinteractions/invariants/validator.go

@@ -0,0 +1,168 @@
+// Package invariants validates normalized D&D NPC interaction artifacts.
+package invariants
+
+import (
+	"context"
+	"fmt"
+	"sort"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	interactionmodel "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcinteractions"
+	npcregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/registry"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
+	interactionshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/shape"
+)
+
+const (
+	Key        = "normalize/dnd/npc-interactions/invariants"
+	ReasonCode = "invalid_npc_interaction_normalization"
+	policy     = "dnd.npc_interactions.validator.normalized.v1"
+)
+
+type Options struct{}
+
+type Validator struct {
+	npcResolver *npcregistry.Resolver
+}
+
+var _ contracts.TypedValidator[dnd.NPCInteractionList] = (*Validator)(nil)
+var _ contracts.ManifestMetadataProvider = (*Validator)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Validator)(nil)
+
+func New(_ Options, references ...contracts.ReferenceSet) (*Validator, error) {
+	if len(references) > 1 {
+		return nil, fmt.Errorf("NPC interaction invariants validator accepts at most one reference set")
+	}
+	var referenceSet contracts.ReferenceSet
+	if len(references) == 1 {
+		referenceSet = references[0]
+	}
+	resolver, err := npcregistry.NewResolver(referenceSet)
+	if err != nil {
+		return nil, fmt.Errorf("prepare NPC registry: %w", err)
+	}
+	return &Validator{npcResolver: resolver}, nil
+}
+
+func (v *Validator) Name() string { return Key }
+func (v *Validator) ExecutionClass() contracts.ExecutionClass {
+	return contracts.ExecutionClassDeterministic
+}
+
+func (v *Validator) ManifestMetadata() map[string]any {
+	if v == nil || v.npcResolver == nil {
+		return nil
+	}
+	metadata := map[string]any{"policy": policy}
+	seeded := v.npcResolver.Seeded()
+	if seeded.Bound() {
+		metadata["npc_registry_digest"] = seeded.Digest()
+		metadata["npc_count"] = seeded.Count()
+	}
+	return metadata
+}
+
+func (v *Validator) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	if v == nil || v.npcResolver == nil {
+		return nil
+	}
+	return []pipeline.CheckpointFingerprint{
+		{Name: "policy", Value: policy},
+		{Name: "npc_registry", Value: v.npcResolver.Seeded().ProjectionDigest()},
+	}
+}
+
+func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationRequest[dnd.NPCInteractionList]) (contracts.ValidationResult, error) {
+	if interactionshape.Validate(req.Value) != nil || !allSourceRefsValid(req.Source, req.Value) {
+		return contracts.ValidationResult{Approved: true}, nil
+	}
+	if v == nil || v.npcResolver == nil {
+		return contracts.ValidationResult{}, fmt.Errorf("NPC interaction invariants validator must not be nil")
+	}
+	npcRegistry, err := v.npcResolver.Resolve(req.References)
+	if err != nil {
+		return contracts.ValidationResult{}, fmt.Errorf("resolve NPC registry: %w", err)
+	}
+	if !npcRegistry.Bound() {
+		return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: "invalid NPC interaction normalization: NPC registry reference is required"}, nil
+	}
+	issues := issuesFor(req.Source, req.Value, npcRegistry)
+	if len(issues) == 0 {
+		return contracts.ValidationResult{Approved: true}, nil
+	}
+	return contracts.ValidationResult{
+		Approved:   false,
+		ReasonCode: ReasonCode,
+		Message:    diagnostics.Aggregate("invalid NPC interaction normalization", issues),
+	}, nil
+}
+
+func allSourceRefsValid(doc *source.SourceDocument, value dnd.NPCInteractionList) bool {
+	for _, interaction := range value.Interactions {
+		if !interactionmodel.ValidSourceRefs(doc, interaction.SourceRefs) {
+			return false
+		}
+	}
+	return true
+}
+
+func issuesFor(doc *source.SourceDocument, value dnd.NPCInteractionList, npcRegistry *npcregistry.Registry) []string {
+	issues := make([]string, 0)
+	for index, interaction := range value.Interactions {
+		prefix := fmt.Sprintf("interactions[%d]", index)
+		if canonical, ok := npcRegistry.Lookup(interaction.Name); ok && interaction.Name != canonical.Name {
+			issues = append(issues, prefix+".name is not the canonical NPC display name: "+diagnostics.Quote(interaction.Name))
+		}
+		for refIndex := 1; refIndex < len(interaction.SourceRefs); refIndex++ {
+			previous := interaction.SourceRefs[refIndex-1]
+			current := interaction.SourceRefs[refIndex]
+			if interactionmodel.SourceRefLess(doc, current, previous) {
+				issues = append(issues, fmt.Sprintf("%s.source_refs are not in canonical order at index %d", prefix, refIndex))
+			} else if current == previous {
+				issues = append(issues, fmt.Sprintf("%s.source_refs[%d] duplicates the previous reference", prefix, refIndex))
+			}
+		}
+	}
+
+	if !sort.SliceIsSorted(value.Interactions, func(left, right int) bool {
+		return interactionmodel.Less(doc, value.Interactions[left], value.Interactions[right])
+	}) {
+		issues = append(issues, "interactions are not in canonical order")
+	}
+	seen := make(map[string]int)
+	for index, interaction := range value.Interactions {
+		key := interactionmodel.ExactIdentity(interaction)
+		if previous, ok := seen[key]; ok {
+			issues = append(issues, fmt.Sprintf("interactions[%d] duplicates interaction %d", index, previous))
+			continue
+		}
+		seen[key] = index
+	}
+	return issues
+}
+
+func Spec() pipeline.ValidatorSpec {
+	return pipeline.ValidatorSpec{Key: Key, ExecutionClass: contracts.ExecutionClassDeterministic}
+}
+
+func Register(registry *pipeline.ValidatorRegistry) error {
+	return pipeline.RegisterTypedValidatorBuilder(registry, dnd.NPCInteractionListKind, Spec(), validateOptions, func(request pipeline.BuildRequest) (contracts.TypedValidator[dnd.NPCInteractionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(options, request.References)
+	})
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, err
+	}
+	return Options{}, nil
+}
+
+func validateOptions(options map[string]any) error { _, err := DecodeOptions(options); return err }

internal/modules/dnd/validate/npcinteractions/invariants/validator_test.go

@@ -0,0 +1,169 @@
+package invariants
+
+import (
+	"context"
+	"encoding/json"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	npccodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcs"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
+	npcregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/registry"
+)
+
+func TestValidatorApprovesCanonicalNormalizedInteractions(t *testing.T) {
+	references := registryReferences(t, "Aria", "Borin")
+	result, err := newValidator(t, references).Validate(context.Background(), request(references, normalizedList()))
+	if err != nil || !result.Approved {
+		t.Fatalf("Validate() = %#v, %v", result, err)
+	}
+}
+
+func TestValidatorRejectsOwnedCanonicalNameReferenceOrderListOrderAndDuplicates(t *testing.T) {
+	references := registryReferences(t, "Aria", "Borin")
+	for _, test := range []struct {
+		name   string
+		mutate func(*dnd.NPCInteractionList)
+		want   string
+	}{
+		{"canonical name", func(value *dnd.NPCInteractionList) { value.Interactions[0].Name = " aria " }, "canonical NPC display name"},
+		{"reference order", func(value *dnd.NPCInteractionList) {
+			value.Interactions[0].SourceRefs = []source.SourceRef{{SourceID: "session", StartUnitID: 20, EndUnitID: 20}, {SourceID: "session", StartUnitID: 10, EndUnitID: 10}}
+		}, "not in canonical order"},
+		{"duplicate reference", func(value *dnd.NPCInteractionList) {
+			value.Interactions[0].SourceRefs = append(value.Interactions[0].SourceRefs, value.Interactions[0].SourceRefs[0])
+		}, "duplicates the previous reference"},
+		{"list order", func(value *dnd.NPCInteractionList) {
+			value.Interactions[0], value.Interactions[1] = value.Interactions[1], value.Interactions[0]
+		}, "interactions are not in canonical order"},
+		{"name tie breaker", func(value *dnd.NPCInteractionList) {
+			value.Interactions[0] = dnd.NPCInteraction{Name: "Borin", Kind: dnd.NPCInteractionKindDialogue, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 10, EndUnitID: 10}}}
+			value.Interactions[1] = dnd.NPCInteraction{Name: "Aria", Kind: dnd.NPCInteractionKindDialogue, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 10, EndUnitID: 10}}}
+		}, "interactions are not in canonical order"},
+		{"duplicate record", func(value *dnd.NPCInteractionList) {
+			value.Interactions = append(value.Interactions, value.Interactions[0])
+		}, "duplicates interaction"},
+	} {
+		t.Run(test.name, func(t *testing.T) {
+			value := normalizedList()
+			test.mutate(&value)
+			result, err := newValidator(t, references).Validate(context.Background(), request(references, value))
+			if err != nil || result.Approved || result.ReasonCode != ReasonCode || !strings.Contains(result.Message, test.want) {
+				t.Fatalf("Validate() = %#v, %v; want %q", result, err, test.want)
+			}
+		})
+	}
+}
+
+func TestValidatorAcceptsValidEvidenceBeforeInvalidEvidence(t *testing.T) {
+	references := registryReferences(t, "Aria", "Borin")
+	value := dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{
+		{Name: "Borin", Kind: dnd.NPCInteractionKindMentioned, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 20, EndUnitID: 20}}},
+		{Name: "Aria", Kind: dnd.NPCInteractionKindDialogue, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 99, EndUnitID: 99}}},
+	}}
+	result, err := newValidator(t, references).Validate(context.Background(), request(references, value))
+	if err != nil || !result.Approved {
+		t.Fatalf("Validate() = %#v, %v", result, err)
+	}
+}
+
+func TestValidatorDefersShapeAndSourceReferenceFailuresAndRequiresRegistry(t *testing.T) {
+	references := registryReferences(t, "Aria", "Borin")
+	for _, value := range []dnd.NPCInteractionList{
+		{Interactions: []dnd.NPCInteraction{{Name: "Aria"}}},
+		{Interactions: []dnd.NPCInteraction{{Name: "Aria", Kind: dnd.NPCInteractionKindDialogue, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 99, EndUnitID: 99}}}}},
+	} {
+		result, err := newValidator(t, references).Validate(context.Background(), request(references, value))
+		if err != nil || !result.Approved {
+			t.Fatalf("deferral = %#v, %v", result, err)
+		}
+	}
+	result, err := newValidator(t).Validate(context.Background(), request(contracts.ReferenceSet{}, normalizedList()))
+	if err != nil || result.Approved || result.ReasonCode != ReasonCode || !strings.Contains(result.Message, "required") {
+		t.Fatalf("unbound registry = %#v, %v", result, err)
+	}
+}
+
+func TestValidatorResolvesGeneratedRegistryAndKeepsMetadataAndInputsImmutable(t *testing.T) {
+	references := registryReferences(t, "Aria", "Borin")
+	validator := newValidator(t)
+	value := normalizedList()
+	before := cloneList(value)
+	result, err := validator.Validate(context.Background(), request(references, value))
+	if err != nil || !result.Approved || !reflect.DeepEqual(value, before) {
+		t.Fatalf("generated validation = %#v, %v; value=%#v", result, err, value)
+	}
+	metadata, err := json.Marshal(newValidator(t, references).ManifestMetadata())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if strings.Contains(string(metadata), "Aria") || strings.Contains(string(metadata), "other-session") {
+		t.Fatalf("metadata leaked registry content: %s", metadata)
+	}
+	if got := newValidator(t, references).CheckpointFingerprints(); len(got) != 2 || got[0].Value != policy || !strings.HasPrefix(got[1].Value, "sha256:") {
+		t.Fatalf("CheckpointFingerprints() = %#v", got)
+	}
+	registry := pipeline.NewValidatorRegistry()
+	if err := Register(registry); err != nil {
+		t.Fatal(err)
+	}
+	if _, err := DecodeOptions(map[string]any{"unexpected": true}); err == nil {
+		t.Fatal("DecodeOptions() accepted unknown option")
+	}
+}
+
+func newValidator(t *testing.T, references ...contracts.ReferenceSet) *Validator {
+	t.Helper()
+	validator, err := New(Options{}, references...)
+	if err != nil {
+		t.Fatal(err)
+	}
+	return validator
+}
+
+func request(references contracts.ReferenceSet, value dnd.NPCInteractionList) contracts.TypedValidationRequest[dnd.NPCInteractionList] {
+	return contracts.TypedValidationRequest[dnd.NPCInteractionList]{Source: document(), References: references, Value: value}
+}
+
+func document() *source.SourceDocument {
+	return &source.SourceDocument{ID: "session", Units: []source.SourceUnit{{ID: 10}, {ID: 20}}}
+}
+
+func normalizedList() dnd.NPCInteractionList {
+	return dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{
+		{Name: "Aria", Kind: dnd.NPCInteractionKindDialogue, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 10, EndUnitID: 10}}},
+		{Name: "Borin", Kind: dnd.NPCInteractionKindMentioned, SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 20, EndUnitID: 20}}},
+	}}
+}
+
+func cloneList(value dnd.NPCInteractionList) dnd.NPCInteractionList {
+	copyValue := dnd.NPCInteractionList{Interactions: make([]dnd.NPCInteraction, len(value.Interactions))}
+	for index, interaction := range value.Interactions {
+		copyValue.Interactions[index] = interaction
+		copyValue.Interactions[index].SourceRefs = append([]source.SourceRef(nil), interaction.SourceRefs...)
+	}
+	return copyValue
+}
+
+func registryReferences(t *testing.T, names ...string) contracts.ReferenceSet {
+	t.Helper()
+	npcs := make([]dnd.NPC, len(names))
+	for index, name := range names {
+		npcs[index] = dnd.NPC{ID: identity.DeriveID(name), Name: name, SourceRefs: []source.SourceRef{{SourceID: "other-session", StartUnitID: index + 1, EndUnitID: index + 1}}}
+	}
+	content, err := npccodec.New().Encode(dnd.NPCList{NPCs: npcs})
+	if err != nil {
+		t.Fatal(err)
+	}
+	return contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		npcregistry.ReferenceSlot: {
+			Slot:  contracts.ReferenceSlot{Name: npcregistry.ReferenceSlot},
+			Items: []contracts.ReferenceItem{{SlotName: npcregistry.ReferenceSlot, MediaType: npccodec.MediaType, Content: content, Origin: contracts.ReferenceOrigin{Type: "generated"}}},
+		},
+	}}
+}

internal/modules/dnd/validate/npcinteractions/registry/validator.go

@@ -0,0 +1,130 @@
+// Package registry validates D&D NPC interaction names against NPC grounding.
+package registry
+
+import (
+	"context"
+	"fmt"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	npcregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/registry"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
+	interactionshape "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/validate/npcinteractions/shape"
+)
+
+const (
+	Key        = "extract/dnd/npc-interactions/registry"
+	ReasonCode = "invalid_npc_interaction_registry"
+	policy     = "dnd.npc_interactions.validator.registry.v1"
+)
+
+type Options struct{}
+
+type Validator struct {
+	npcResolver *npcregistry.Resolver
+}
+
+var _ contracts.TypedValidator[dnd.NPCInteractionList] = (*Validator)(nil)
+var _ contracts.ManifestMetadataProvider = (*Validator)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Validator)(nil)
+
+func New(_ Options, references ...contracts.ReferenceSet) (*Validator, error) {
+	if len(references) > 1 {
+		return nil, fmt.Errorf("NPC interaction registry validator accepts at most one reference set")
+	}
+	var referenceSet contracts.ReferenceSet
+	if len(references) == 1 {
+		referenceSet = references[0]
+	}
+	resolver, err := npcregistry.NewResolver(referenceSet)
+	if err != nil {
+		return nil, fmt.Errorf("prepare NPC registry: %w", err)
+	}
+	return &Validator{npcResolver: resolver}, nil
+}
+
+func (v *Validator) Name() string { return Key }
+func (v *Validator) ExecutionClass() contracts.ExecutionClass {
+	return contracts.ExecutionClassDeterministic
+}
+
+func (v *Validator) ManifestMetadata() map[string]any {
+	if v == nil || v.npcResolver == nil {
+		return nil
+	}
+	metadata := map[string]any{"policy": policy}
+	seeded := v.npcResolver.Seeded()
+	if seeded.Bound() {
+		metadata["npc_registry_digest"] = seeded.Digest()
+		metadata["npc_count"] = seeded.Count()
+	}
+	return metadata
+}
+
+func (v *Validator) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	if v == nil || v.npcResolver == nil {
+		return nil
+	}
+	return []pipeline.CheckpointFingerprint{
+		{Name: "policy", Value: policy},
+		{Name: "npc_registry", Value: v.npcResolver.Seeded().ProjectionDigest()},
+	}
+}
+
+func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationRequest[dnd.NPCInteractionList]) (contracts.ValidationResult, error) {
+	if interactionshape.Validate(req.Value) != nil {
+		return contracts.ValidationResult{Approved: true}, nil
+	}
+	if v == nil || v.npcResolver == nil {
+		return contracts.ValidationResult{}, fmt.Errorf("NPC interaction registry validator must not be nil")
+	}
+	npcRegistry, err := v.npcResolver.Resolve(req.References)
+	if err != nil {
+		return contracts.ValidationResult{}, fmt.Errorf("resolve NPC registry: %w", err)
+	}
+	if !npcRegistry.Bound() {
+		return rejection([]string{"NPC registry reference is required"}), nil
+	}
+	issues := make([]string, 0)
+	for index, interaction := range req.Value.Interactions {
+		if _, ok := npcRegistry.Lookup(interaction.Name); !ok {
+			issues = append(issues, fmt.Sprintf("interactions[%d].name is not in the NPC registry: %s", index, diagnostics.Quote(interaction.Name)))
+		}
+	}
+	if len(issues) == 0 {
+		return contracts.ValidationResult{Approved: true}, nil
+	}
+	return rejection(issues), nil
+}
+
+func rejection(issues []string) contracts.ValidationResult {
+	return contracts.ValidationResult{
+		Approved:   false,
+		ReasonCode: ReasonCode,
+		Message:    diagnostics.Aggregate("invalid NPC interaction registry", issues),
+	}
+}
+
+func Spec() pipeline.ValidatorSpec {
+	return pipeline.ValidatorSpec{Key: Key, ExecutionClass: contracts.ExecutionClassDeterministic}
+}
+
+func Register(registry *pipeline.ValidatorRegistry) error {
+	return pipeline.RegisterTypedValidatorBuilder(registry, dnd.NPCInteractionListKind, Spec(), validateOptions, func(request pipeline.BuildRequest) (contracts.TypedValidator[dnd.NPCInteractionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(options, request.References)
+	})
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, err
+	}
+	return Options{}, nil
+}
+
+func validateOptions(options map[string]any) error { _, err := DecodeOptions(options); return err }

internal/modules/dnd/validate/npcinteractions/registry/validator_test.go

@@ -0,0 +1,137 @@
+package registry
+
+import (
+	"context"
+	"encoding/json"
+	"reflect"
+	"strings"
+	"testing"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/core/source"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	npccodec "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/codec/npcs"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/identity"
+	npcregistry "gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/npcs/registry"
+)
+
+func TestValidatorRecognizesRegistryNamesAndRejectsUnknownNames(t *testing.T) {
+	references := registryReferences(t, "Mira Thorn")
+	validator := newValidator(t, references)
+	value := validList("  mira thorn  ")
+	result, err := validator.Validate(context.Background(), request(references, value))
+	if err != nil || !result.Approved {
+		t.Fatalf("recognized result = %#v, %v", result, err)
+	}
+	value.Interactions[0].Name = "Unknown NPC"
+	result, err = validator.Validate(context.Background(), request(references, value))
+	if err != nil || result.Approved || result.ReasonCode != ReasonCode || !strings.Contains(result.Message, "interactions[0].name") {
+		t.Fatalf("unknown result = %#v, %v", result, err)
+	}
+}
+
+func TestValidatorRequiresRegistryAndResolvesGeneratedReferenceAtOperationTime(t *testing.T) {
+	validator := newValidator(t)
+	value := validList("Mira Thorn")
+	result, err := validator.Validate(context.Background(), request(contracts.ReferenceSet{}, value))
+	if err != nil || result.Approved || result.ReasonCode != ReasonCode || !strings.Contains(result.Message, "required") {
+		t.Fatalf("unbound result = %#v, %v", result, err)
+	}
+	references := registryReferences(t, "Mira Thorn")
+	result, err = validator.Validate(context.Background(), request(references, value))
+	if err != nil || !result.Approved {
+		t.Fatalf("generated result = %#v, %v", result, err)
+	}
+
+	empty := registryReferences(t)
+	result, err = newValidator(t, empty).Validate(context.Background(), request(empty, dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{}}))
+	if err != nil || !result.Approved {
+		t.Fatalf("empty registry result = %#v, %v", result, err)
+	}
+}
+
+func TestValidatorRejectsMalformedRegistryWithoutContentAndKeepsMetadataSafe(t *testing.T) {
+	malformed := contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		npcregistry.ReferenceSlot: {
+			Slot:  contracts.ReferenceSlot{Name: npcregistry.ReferenceSlot},
+			Items: []contracts.ReferenceItem{{SlotName: npcregistry.ReferenceSlot, MediaType: "application/json", Content: []byte(`{"secret":"private source"}`)}},
+		},
+	}}
+	if _, err := New(Options{}, malformed); err == nil || strings.Contains(err.Error(), "private source") {
+		t.Fatalf("New() error = %v", err)
+	}
+	references := registryReferences(t, "Mira Thorn")
+	validator := newValidator(t, references)
+	metadata, err := json.Marshal(validator.ManifestMetadata())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if strings.Contains(string(metadata), "Mira Thorn") || strings.Contains(string(metadata), "other-session") {
+		t.Fatalf("metadata leaked registry content: %s", metadata)
+	}
+	if got := validator.CheckpointFingerprints(); len(got) != 2 || got[0].Value != policy || !strings.HasPrefix(got[1].Value, "sha256:") {
+		t.Fatalf("CheckpointFingerprints() = %#v", got)
+	}
+}
+
+func TestValidatorDefersShapeAndDoesNotMutateOrMisregister(t *testing.T) {
+	references := registryReferences(t, "Mira Thorn")
+	validator := newValidator(t, references)
+	malformed := dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{{Name: "Mira Thorn"}}}
+	result, err := validator.Validate(context.Background(), request(references, malformed))
+	if err != nil || !result.Approved {
+		t.Fatalf("shape deferral = %#v, %v", result, err)
+	}
+	value := validList("Mira Thorn")
+	before := value
+	_, err = validator.Validate(context.Background(), request(references, value))
+	if err != nil || !reflect.DeepEqual(value, before) {
+		t.Fatalf("Validate() mutated value: %#v", value)
+	}
+	registry := pipeline.NewValidatorRegistry()
+	if err := Register(registry); err != nil {
+		t.Fatal(err)
+	}
+	if _, err := DecodeOptions(map[string]any{"unexpected": true}); err == nil {
+		t.Fatal("DecodeOptions() accepted unknown option")
+	}
+}
+
+func newValidator(t *testing.T, references ...contracts.ReferenceSet) *Validator {
+	t.Helper()
+	validator, err := New(Options{}, references...)
+	if err != nil {
+		t.Fatal(err)
+	}
+	return validator
+}
+
+func request(references contracts.ReferenceSet, value dnd.NPCInteractionList) contracts.TypedValidationRequest[dnd.NPCInteractionList] {
+	return contracts.TypedValidationRequest[dnd.NPCInteractionList]{References: references, Value: value}
+}
+
+func validList(name string) dnd.NPCInteractionList {
+	return dnd.NPCInteractionList{Interactions: []dnd.NPCInteraction{{
+		Name: name, Kind: dnd.NPCInteractionKindDialogue,
+		SourceRefs: []source.SourceRef{{SourceID: "session", StartUnitID: 1, EndUnitID: 1}},
+	}}}
+}
+
+func registryReferences(t *testing.T, names ...string) contracts.ReferenceSet {
+	t.Helper()
+	npcs := make([]dnd.NPC, len(names))
+	for index, name := range names {
+		npcs[index] = dnd.NPC{ID: identity.DeriveID(name), Name: name, SourceRefs: []source.SourceRef{{SourceID: "other-session", StartUnitID: index + 1, EndUnitID: index + 1}}}
+	}
+	content, err := npccodec.New().Encode(dnd.NPCList{NPCs: npcs})
+	if err != nil {
+		t.Fatal(err)
+	}
+	return contracts.ReferenceSet{Slots: map[string]contracts.ResolvedReferenceSlot{
+		npcregistry.ReferenceSlot: {
+			Slot:  contracts.ReferenceSlot{Name: npcregistry.ReferenceSlot},
+			Items: []contracts.ReferenceItem{{SlotName: npcregistry.ReferenceSlot, MediaType: npccodec.MediaType, Content: content, Origin: contracts.ReferenceOrigin{Type: "generated"}}},
+		},
+	}}
+}

internal/modules/dnd/validate/npcinteractions/shape/validator.go

@@ -0,0 +1,106 @@
+// Package shape validates required D&D NPC interaction candidate fields.
+package shape
+
+import (
+	"context"
+	"fmt"
+	"strings"
+
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/contracts"
+	"gitea.maximumdirect.net/eric/notarius/internal/framework/pipeline"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd"
+	"gitea.maximumdirect.net/eric/notarius/internal/modules/dnd/shared/diagnostics"
+)
+
+const (
+	Key        = "extract/dnd/npc-interactions/shape"
+	ReasonCode = "invalid_npc_interaction_shape"
+	policy     = "dnd.npc_interactions.validator.shape.v1"
+)
+
+type Options struct{}
+type Validator struct{}
+
+var _ contracts.TypedValidator[dnd.NPCInteractionList] = (*Validator)(nil)
+var _ pipeline.CheckpointFingerprintProvider = (*Validator)(nil)
+
+func New(Options) *Validator      { return &Validator{} }
+func (v *Validator) Name() string { return Key }
+func (v *Validator) ExecutionClass() contracts.ExecutionClass {
+	return contracts.ExecutionClassDeterministic
+}
+func (v *Validator) CheckpointFingerprints() []pipeline.CheckpointFingerprint {
+	return []pipeline.CheckpointFingerprint{{Name: "policy", Value: policy}}
+}
+
+func (v *Validator) Validate(_ context.Context, req contracts.TypedValidationRequest[dnd.NPCInteractionList]) (contracts.ValidationResult, error) {
+	if err := Validate(req.Value); err != nil {
+		return contracts.ValidationResult{Approved: false, ReasonCode: ReasonCode, Message: err.Error()}, nil
+	}
+	return contracts.ValidationResult{Approved: true}, nil
+}
+
+func Validate(value dnd.NPCInteractionList) error {
+	issues := issuesFor(value)
+	if len(issues) == 0 {
+		return nil
+	}
+	return fmt.Errorf("%s", diagnostics.Aggregate("invalid NPC interaction shape", issues))
+}
+
+func issuesFor(value dnd.NPCInteractionList) []string {
+	if value.Interactions == nil {
+		return []string{"interactions must be present"}
+	}
+	issues := make([]string, 0)
+	for index, interaction := range value.Interactions {
+		prefix := fmt.Sprintf("interactions[%d]", index)
+		if strings.TrimSpace(interaction.Name) == "" {
+			issues = append(issues, prefix+".name must not be empty: "+diagnostics.Quote(interaction.Name))
+		}
+		if !validKind(interaction.Kind) {
+			issues = append(issues, prefix+".kind is unsupported: "+diagnostics.Quote(string(interaction.Kind)))
+		}
+		if len(interaction.SourceRefs) == 0 {
+			issues = append(issues, prefix+".source_refs must contain at least one reference")
+		}
+	}
+	return issues
+}
+
+func validKind(value dnd.NPCInteractionKind) bool {
+	switch value {
+	case dnd.NPCInteractionKindMentioned,
+		dnd.NPCInteractionKindNoncombatPresence,
+		dnd.NPCInteractionKindDialogue,
+		dnd.NPCInteractionKindCombatAlly,
+		dnd.NPCInteractionKindCombatOpponent,
+		dnd.NPCInteractionKindOther:
+		return true
+	default:
+		return false
+	}
+}
+
+func Spec() pipeline.ValidatorSpec {
+	return pipeline.ValidatorSpec{Key: Key, ExecutionClass: contracts.ExecutionClassDeterministic}
+}
+
+func Register(registry *pipeline.ValidatorRegistry) error {
+	return pipeline.RegisterTypedValidatorBuilder(registry, dnd.NPCInteractionListKind, Spec(), validateOptions, func(request pipeline.BuildRequest) (contracts.TypedValidator[dnd.NPCInteractionList], error) {
+		options, err := DecodeOptions(request.Options)
+		if err != nil {
+			return nil, err
+		}
+		return New(options), nil
+	})
+}
+
+func DecodeOptions(options map[string]any) (Options, error) {
+	if err := pipeline.RejectUnknownOptions(options); err != nil {
+		return Options{}, err
+	}
+	return Options{}, nil
+}
+
+func validateOptions(options map[string]any) error { _, err := DecodeOptions(options); return err }