Add standalone D&D combat turn extraction

This commit is contained in:
2026-07-21 05:08:02 +00:00
parent 6dd695611c
commit 2fbb3813aa
22 changed files with 1924 additions and 4 deletions

View File

@@ -1,8 +1,8 @@
# D&D Combat-Turn Artifact Contract
This document defines the durable artifact and serialization boundary for
D&D combat turns. It does not define extraction, validation beyond the codec's
structural checks, normalization, prompts, or a selectable pipeline lane.
This document defines the durable artifact, serialization, extraction, and
candidate-validation boundaries for D&D combat turns. Normalization and a
selectable production pipeline lane are not part of this contract yet.
## Artifact identity
@@ -71,3 +71,53 @@ The codec owns the durable JSON Schema, whose object layers all set
The maintained compact fixture is
`internal/modules/dnd/codec/combatturns/testdata/dnd_combat_turns.v1.json`.
## Extraction boundary
The standalone extractor uses these identities:
| Property | Value |
| --- | --- |
| Extractor key | `dnd/combat-turns` |
| Capability | `dnd.combat_turns` |
| Prompt ID | `dnd.combat_turns` |
| Prompt version | `v1` |
| Private response-schema key | `dnd_combat_turns_llm` |
| Private response-schema ID | `notarius.dnd.combat_turns.llm` |
| Default profile | `gemini-2-flash` |
It requires `chunks` and `source.transcript`, accepts no options, and makes one
structured completion for each supplied chunk. The prompt receives the
chunk-scoped transcript plus the existing `players`, `party`, and `glossary`
inputs, and optionally the deprecated `roster` reference through the shared
party mapping. The optional `npcs` reference is an approved normalized NPC
artifact used only for identity grounding; it never supplies combat evidence.
The private response shape is the same as the durable turn/action shape except
that source references contain only `start_unit_id` and `end_unit_id`. The
extractor assigns the current source ID, removes exact duplicate ranges, and
stable-sorts turns by the earliest valid source-document position. Numeric unit
IDs are identifiers; source-document slice position determines chronology.
Malformed candidate fields remain in the typed result for deterministic
validators to report.
## Deterministic candidate validation
The standalone validator keys are:
| Validator | Responsibility |
| --- | --- |
| `extract/dnd/combat-turns/shape` | Required arrays, strings, nullable fields, positive rounds, and supported enum values. |
| `extract/dnd/combat-turns/source_refs` | Source identity, source-unit existence, and range order through the source document. |
| `extract/dnd/combat-turns/source_relatedness` | At most one advisory warning per turn when the actor or declared action is not related to cited transcript text. |
Source-reference and relatedness validators defer malformed shape to the shape
validator. Relatedness also defers when any cited source range is invalid. It
combines overlapping cited ranges once in document order, compares actors with
the shared Unicode-aware NPC identity policy, and checks declaration tokens of
at least four Unicode code points against complete cited-text tokens. Targets
are not checked deterministically.
The extractor and validators are package-complete but are not registered by the
production D&D family registrar yet. Normalization, production composition,
and selectable configuration are defined when implemented.