100%

Verification: pytest + L12 production-path on aims-burden-tracking

Goal

Prove the whole forensic mode works end-to-end, including the L12 production entry point on the REAL motivating plan. The original failure must be demonstrably impossible: a /continue-roadmap on a zeroed plan halts until forensic /review-plan runs; forensic /review-plan on plans/aims-burden-tracking produces the dossier (132 dead-ends, 17 evolutions, OBE history), fires /review-section across all 23 sections, applies reshuffles, and sets forensic_reviewed.

Implementation Sketch

Fix Completeness (CLAUDE.md): production-path L12 end-to-end on the real entry point (not seam-injected); full pytest green across touched suites; /review-plan self-review + /impl-hygiene-review. This plan is tooling-scope (scripts + skills) so compiler dual-exec parity is N/A; pytest + the L12 real-run are the gates.

Spec References

scripts/review_plan_runtime/tests/; scripts/plan_orchestrator/tests/; scripts/plan_corpus/tests/** (touched); plans/aims-burden-tracking (the L12 target); /review-plan (self-review); /impl-hygiene-review.

Work Items

  • Full pytest green: scripts/review_plan_runtime/tests + scripts/plan_orchestrator/tests + scripts/plan_corpus/tests for every touched module; no skips without a tracked anchor.
  • L12 production-path end-to-end: run forensic /review-plan on the REAL plans/aims-burden-tracking (zeroed); assert the committed .review-plan/ dossier enumerates the dead-ends + refuted evolutions + OBE/restructure history, the all-section fan-out fired across every section, reshuffles were applied + logged, and forensic_reviewed flipped true.
  • Recurrence guard: a /continue-roadmap —autopilot on a zeroed plan HALTS at forensic_review_required until forensic /review-plan runs — the exact original failure is structurally impossible. Pin it as a test.
  • /review-plan self-review of this plan (the improved skill on its own plan) + /impl-hygiene-review; resolve findings inline.

Fresh intel (regenerated)

{ “schema_version”: 1, “target”: { “kind”: “plan-section”, “ref”: “zeroed-plan-forensic-review/s-c0e7168a” }, “generated_at”: “2026-06-20T04:44:56.780613+00:00”, “graph_state”: { “head_sha”: “bd1ba13c”, “last_code_import_at”: “2026-06-20T04:42:40.628Z”, “embedding_stale”: false, “insights_stale”: false, “cpg_stale”: false }, “surfaces”: {}, “summary”: “intel-package[plan-section:zeroed-plan-forensic-review/s-c0e7168a] agent-authored dossier”, “degraded”: false, “dossier”: { “objective_symbols”: [], “tiers”: {}, “agent_authored”: true, “difficulty”: { “class”: “routine”, “research_online”: false, “research_mode”: “auto”, “signals”: [], “plan_dir”: “/home/eric/projects/ori_lang/plans/zeroed-plan-forensic-review” } } }

(full dossier: verification-end-to-end—s-c0e7168a.intel.json)