audita/docs/architecture.md

Audita Go Architecture

Scope and intent

This document describes:

• the current implemented Go architecture; and
• the intended final architecture for later rewrite phases.

Status labels are explicit so future engineers and LLM agents do not assume unimplemented behavior exists.

Current implementation status

Implemented today:

• Go CLI entrypoint and audita process wiring.
• Config defaults, env loading, CLI override precedence, and validation.
• Transcript and glossary parsing/validation.
• Deterministic transcript normalization.
• Deterministic token estimation and transcript chunking.
• Per-run diagnostics directory creation plus Phase 6 process-level artifacts.
• Process report JSON output with diagnostics artifact references.
• Framework foundation packages for contracts and proposal application.
• Production runner orchestration package with deterministic sequential module execution.
• Module-level report structures with applied/skipped change records.
• Runtime validator models and deterministic validators.
• Deterministic validator-chain execution in the runner with cardinality enforcement.
• Module-level validator decision/rejection reporting.
• Internal structured LLM client contract plus an instructor-go-backed adapter package.
• Bounded LLM scheduler/semaphore infrastructure with context-aware permit handling.
• Runtime primary/validation LLM effective-config resolution helpers with validation inheritance.
• Generic JSON prompt/response diagnostics writer primitives with secret redaction.
• LLM-backed validator models, prompt builders, batching, and runtime execution.
• Runner wiring for LLM validators via the internal structured LLM abstraction and scheduler hooks.
• LLM validator diagnostics artifacts and report-level decision metadata paths.

Not implemented in CLI runtime path today:

• Real module execution pipeline (glossary, homophones, spoken_word, grammar).
• Structured LLM proposal generation.
• Shared module proposal-generation framework and module registry for real modules.
• End-to-end transcript polishing with real module behavior.

Phase sequencing note:

• Phase 9 LLM infrastructure is complete (structured client, scheduler, effective config resolution, diagnostics primitives);
• Phase 10 LLM-backed validator runtime integration is complete;
• shared module proposal-generation and module registry work remain Phase 11.

Actual Go package layout

cmd/audita/
  main.go

internal/cli/
  run.go

internal/core/config/
  config.go
  env.go
  flags.go
  redaction.go
  validation.go

internal/core/schema/
  transcript.go
  glossary.go
  errors.go

internal/core/io/
  files.go

internal/core/normalization/
  normalize.go
  tokens.go

internal/core/chunking/
  sections.go
  summary.go
  tokens.go

internal/core/diagnostics/
  run_dir.go

internal/core/reporting/
  report.go

internal/framework/contracts/
  contracts.go

internal/framework/proposals/
  proposal.go
  policy.go
  preview.go
  apply.go

internal/framework/runner/
  runner.go

internal/framework/validators/
  models.go
  deterministic.go
  llm_models.go
  llm_prompt_builders.go
  llm_batching.go
  llm_validators.go

internal/framework/llm/
  instructor_client.go
  scheduler.go
  effective_config.go
  diagnostics.go

Current CLI behavior

Primary command:

audita process <transcript.json> --glossary <glossary.yaml> [flags]

Current runtime flow (internal/cli/run.go):

1. Load config from env.
2. Parse flags and apply CLI overrides.
3. Validate transcript positional argument and required --glossary.
4. Create per-run diagnostics directory.
5. Read transcript and glossary files.
6. Parse/validate transcript and glossary.
7. Write source transcript artifacts.
8. Normalize transcript.
9. Write normalized transcript and normalization summary artifacts.
10. Chunk normalized transcript and compute chunk summaries.
11. Write chunking summary artifact.
12. Optionally execute runner modules sequentially when an injected module registry/factory is available (used by deterministic tests today).
13. Output working transcript to --output file or stdout.
14. Build process report (phase currently set to phase10-llm-validators).
15. Optionally write --report-json; always write run-dir report.json.
16. Apply work-dir retention.

Important behavior details:

• Glossary is validated but not yet used for real correction module logic.
• Default production CLI behavior remains deterministic normalization/chunking output because no real module implementations are registered yet.
• No real LLM calls occur in the default production runtime path because no real modules are registered yet.
• Success path is generally quiet on stderr.
• Source IDs are preserved into a canonical transcript before normalization; normalization then reassigns output IDs sequentially from 1.

Implemented data contracts

Transcript input

Accepted top-level forms:

• bare JSON array of segments
• object with segments array

Source segment contract:

• id optional integer
• speaker non-empty string
• start finite non-negative number
• end finite non-negative number with end >= start
• text non-empty string
• categories optional array of non-empty strings

Additional checks:

• duplicate explicit source IDs are rejected.

Transcript output

Current output uses schema.TranscriptToJSON and is a bare JSON array of normalized segments:

• id, speaker, start, end, text, optional categories.

Glossary input

YAML with glossary entries. Required fields per entry:

• name, category, summary

Optional:

• aliases, plural

Implemented config/env/flag behavior

Precedence:

1. defaults (config.Default())
2. environment (config.LoadFromEnv())
3. CLI flags (ApplyCLIOverrides)

Implemented config surfaces include:

• module list
• primary and validation LLM settings
• section token controls and target sections
• confidence thresholds
• normalization controls
• work-dir and retention mode

Current caveat:

• LLM/module-related settings are mostly infrastructure-only today; runtime path does not execute LLM or modules.

Implemented structured LLM infrastructure

internal/framework/contracts now defines a typed structured-completion contract:

• StructuredLLMClient.CompleteStructured(ctx, req, out)
• caller-owned typed decode target via out pointer.

internal/framework/llm provides InstructorClient, an internal adapter over github.com/jxnl/instructor-go:

• configurable base_url, model, optional API key, retries, mode, HTTP client, and request timeout;
• OpenAI-compatible endpoint behavior (for example OpenAI/OpenRouter/local-compatible base URLs);
• default mode is JSON mode (ModeJSON), with optional tool-call mode (ModeToolCall);
• request message translation from contracts.LLMMessage to chat-completions messages;
• response metadata mapping (provider/model/token usage) into Audita-owned response types;
• API-key redaction in adapter-returned errors.

Current runtime boundary:

• the default CLI runtime path still does not instantiate real production modules, so no default end-to-end LLM polishing occurs.
• LLM calls are exercised only when test/injected modules and validators are provided.

internal/framework/llm also provides:

• a bounded Scheduler for controlled concurrent LLM calls with reliable permit release;
• primary/validation effective-config resolution helpers, including validation inheritance fallback to primary settings;
• generic interaction diagnostics primitives that write machine-readable JSON artifacts for request metadata, request payload, response payload, and optional error payload with secret redaction.

Implemented normalization behavior

Normalization (internal/core/normalization) currently:

• sorts by segment start time;
• merges adjacent same-speaker segments when constraints pass;
• uses gap-based joiners:
  • gap < ellipsis_gap -> single space join
  • gap >= ellipsis_gap -> ... join
• enforces merged duration and token-limit constraints;
• reassigns output IDs sequentially from 1;
• returns NormalizationSummary with merge and skip counters.

Note: merged categories are concatenated (not deduplicated).

Implemented chunking behavior

Chunking (internal/core/chunking) currently provides:

• deterministic heuristic token estimation;
• contiguous sectioning with section metadata;
• max/min section token validation;
• optional target_sections handling with target-aware merge/split logic;
• summary and detailed summary generation.

Current behavior details:

• if a single segment exceeds max tokens, it is emitted as its own section (not hard-failed);
• section balancing is deterministic but heuristic.

Implemented proposal/replacement infrastructure

internal/framework/proposals provides deterministic foundation logic:

• CorrectionProposal and EnrichedCorrectionProposal models;
• replacement policies: require_unique, replace_all;
• safe preview (PreviewProposalForSegment) with stable skip reasons;
• deterministic apply (ApplyProposals) in ascending proposal_index order;
• applied/skipped change records suitable for reporting.

internal/framework/contracts provides interfaces and run-spec metadata scaffolding, including deterministic repeated module instance naming (ResolveModuleRunSpecs).

These primitives are wired into the production runner and report model. Real module implementations are still pending.

Implemented validator runtime infrastructure

internal/framework/validators provides deterministic validator infrastructure:

• runtime validation request/result models;
• stable validator reason codes;
• cardinality enforcement for validator decisions:
  • missing proposal indexes fail
  • duplicate proposal indexes fail
  • unknown proposal indexes fail
• deterministic validators:
  • confidence threshold by module key/config threshold
  • original-text presence against current working transcript
  • non-empty corrected text
  • identical/no-effect rejection
  • conservative protected glossary-term guard for non-glossary modules

internal/framework/runner executes validator chains in order for each module and applies only validator-approved proposals. Validator rejections are reported distinctly from proposal-application skips.

Implemented LLM-backed validator infrastructure

internal/framework/validators now includes LLM-backed validator support:

• typed request/response models for structured LLM validation;
• prompt builders for:
  • spoken-form plausibility
  • meaning reversal detection
  • editorial review
  • grammar review
  • spoken-word review
• deterministic batching by validation_max_prompt_tokens;
• strict cardinality validation of structured LLM decisions (missing/duplicate/unknown indexes fail);
• safe failure behavior for malformed/invalid structured responses.

internal/framework/runner wires LLM validators into existing validator chains using:

• the internal structured LLM client abstraction (contracts.StructuredLLMClient);
• bounded scheduler hooks for validator call execution;
• diagnostics writer hooks for machine-readable prompt/response artifacts with secret redaction.

Reports and diagnostics (implemented)

Current per-run artifacts include:

• source-transcript.json
• source-transcript-parsed.json
• normalized-transcript.json
• normalization-summary.json
• chunking-summary.json
• invocation.json
• effective-config.json (redacted credentials)
• report.json
• error.log on failure

--report-json writes a separate report file when requested.

Current process reports include diagnostics metadata references for:

• diagnostics directory path;
• source transcript artifact path;
• parsed source transcript artifact path;
• normalized transcript artifact path;
• normalization summary artifact path;
• chunking summary artifact path;
• invocation metadata artifact path;
• redacted effective-config artifact path;
• error-log artifact path on failure.

Current process reports also include:

• module-level results (when runner modules execute), including applied/skipped proposal changes;
• run-level module summary totals and failed module instance metadata.
• module-level validator decisions and validator rejections.
• optional decision-level diagnostic artifact paths for validator LLM interactions when available.

Retention modes implemented in ApplyRetention:

• always: keep all run directories.
• never: keep successful run directories.
• auto: keep failed runs and successful runs with skipped corrections.
• failed runs are always retained.

Current runtime note:

• real module execution is not implemented yet, so normal successful runs generally have no skipped corrections and auto typically removes clean successful run directories.

Intentionally deferred to module/LLM phases:

• module proposal-generation prompt/response diagnostics remain tied to later real-module phases.

Current tests and quality posture

Implemented tests currently cover:

• CLI argument handling and behavior (internal/cli/run_test.go)
• subprocess stdout/stderr and exit-code behavior (cmd/audita/main_integration_test.go)
• config/env/override validation (internal/core/config/*_test.go)
• transcript and glossary schema validation (internal/core/schema/*_test.go)
• deterministic normalization (internal/core/normalization/*_test.go)
• deterministic chunking and summaries (internal/core/chunking/*_test.go)
• proposal preview/apply semantics (internal/framework/proposals/*_test.go)
• contracts/foundation composition tests (internal/framework/contracts/*_test.go)
• runner sequencing and failure behavior with deterministic fake modules (internal/framework/runner/*_test.go)
• CLI runner integration through injected fake module factories (internal/cli/run_test.go)
• validator models, cardinality enforcement, and deterministic validators (internal/framework/validators/*_test.go)
• LLM-backed validator batching, prompt builders, structured-response safety, scheduler hooks, and diagnostics redaction (internal/framework/validators/*_test.go, internal/framework/runner/*_test.go)

Not covered yet (because not implemented): shared module proposal generation, real module implementations, and full transcript-polishing runtime behavior.

Intended final architecture (not yet implemented)

The intended end-state still matches the rewrite plan:

• sequential module pipeline over a mutable working transcript
• real module implementations (glossary, homophones, spoken_word, grammar)
• structured LLM proposal generation
• deterministic and LLM validators
• validator cardinality enforcement in pipeline execution
• proposal application integrated per module stage
• prompt/response diagnostics for LLM/module stages

Until those phases are implemented, documentation and external descriptions should treat the current Go CLI as deterministic preprocessing/reporting infrastructure, not a full LLM transcript polisher.