SuperBizAgent-java/openspec/changes/archive/2026-07-08-executor-eval-audit-closure/tasks.md

1. Eval Contract

• 1.1 Extend DiagnosisEvalCase with optional V2 audit expectations.
• 1.2 Extend DiagnosisEvalResult and report output with observed audit signals.
• 1.3 Add deterministic evaluator checks for Gatekeeper, claim checks, Composer output, and raw JSON leakage.

2. Fixtures and Baseline

• 2.1 Add fixed eval cases for Gatekeeper failure, unsupported claim filtering, and Composer fallback leakage prevention.
• 2.2 Add matching trace fixtures for each new case.
• 2.3 Regenerate baseline JSON and Markdown reports.

3. Documentation

• 3.1 Update eval README with stage 5 scope and verification commands.
• 3.2 Rewrite eval schema documentation so V2 audit fields and case expectations are readable.

4. Verification and Archive

• 4.1 Add or update unit tests for the new V2 audit checks.
• 4.2 Run focused evaluator and agent-chain tests.
• 4.3 Validate and archive the OpenSpec change.
• 4.4 Commit the completed phase 5 changes.