Surfaced by today's untracked-files audit. None of these are accidents —
multiple are referenced by name in CLAUDE.md and memory files but were
never added.
Categories:
- docs/PHASE_AUDIT_GUIDE.md (106 LOC) — Claude Code phase audit guidance
- ops/systemd/lakehouse-langfuse-bridge.service — Langfuse bridge unit
- package.json — top-level npm manifest
- scripts/e2e_pipeline_check.sh + production_smoke.sh — real test scripts
- reports/kimi/audit-last-week*.md — the "Two reports live" CLAUDE.md cites
- tests/multi-agent/scenarios/ — 44 staffing scenarios (cutover decision A)
- tests/multi-agent/playbooks/ — 102 playbook records
- tests/battery/, tests/agent_test/PRD.md, tests/real-world/* — real tests
- sidecar/sidecar/{lab_ui,pipeline_lab}.py — 888 LOC dev-only UIs that
remain in service post-sidecar-drop (commit ba928b1 explicitly kept them)
Sensitivity check: scenarios use synthetic company names ("Heritage Foods",
"Cornerstone Fabrication"); audit reports describe code findings only;
no PII or secrets surfaced.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
4.3 KiB
4.3 KiB
Scenario retrospective — Heritage Foods, 2026-04-21
Executor: qwen3.5:latest Reviewer: qwen3:latest Draft: qwen2.5:latest Overview(T3): gpt-oss:20b
Prior lessons loaded into executor context: 0 (baseline — no prior T3 history)
Events
| At | Kind | Role / Count | Pool | Fills | Turns | Dur(s) | Cites | Gaps |
|---|---|---|---|---|---|---|---|---|
| 10:30 | baseline_fill | Machine Operator × 2 | 988 | ✓ 2 | 2 | 20.8 | 0 | 2 |
| 10:00 | recurring | Receiving Clerk × 1 | - | ✗ 0 | 0 | 18.2 | 0 | 1 |
| 09:30 | expansion | Packer × 3 | - | ✗ 0 | 0 | 18.5 | 0 | 1 |
| 14:30 | emergency | Assembler × 4 | 571 | ✓ 4 | 2 | 19.4 | 0 | 5 |
| 12:00 | misplacement | Warehouse Associate × 1 | 841 | ✓ 1 | 3 | 15.0 | 0 | 2 |
Final roster
| Worker | Booked | Role | City, ST | Status |
|---|---|---|---|---|
| undefined Amy S. Jenkins | 10:30 | Machine Operator | Indianapolis, IN | no_show |
| undefined Christina Watson | 10:30 | Machine Operator | Indianapolis, IN | confirmed |
| undefined Katherine Q. Gutierrez | 14:30 | Assembler | Indianapolis, IN | confirmed |
| undefined Gary O. Stewart | 14:30 | Assembler | Indianapolis, IN | confirmed |
| undefined Dorothy D. Lewis | 14:30 | Assembler | Indianapolis, IN | confirmed |
| undefined Lauren F. Moore | 14:30 | Assembler | Indianapolis, IN | confirmed |
| undefined Susan X. Ruiz | 12:00 | Warehouse Associate | Indianapolis, IN | confirmed |
Gap signals
double_book
- 10:30 — undefined Christina Watson already booked for 10:30
- 14:30 — undefined Katherine Q. Gutierrez already booked for 10:30
- 14:30 — undefined Gary O. Stewart already booked for 10:30
- 14:30 — undefined Dorothy D. Lewis already booked for 10:30
- 14:30 — undefined Lauren F. Moore already booked for 10:30
- 12:00 — undefined Susan X. Ruiz already booked for 10:30
artifact
- 10:30 — undefined is not an object (evaluating 'f.reason.slice')
- 14:30 — undefined is not an object (evaluating 'f.reason.slice')
- 12:00 — undefined is not an object (evaluating 'f.reason.slice')
drift_or_tool
- 10:00 — aborted — 3 consecutive drift flags
- 09:30 — aborted — 3 consecutive drift flags
fairness
- cross-event — Amy S. Jenkins (undefined) booked 6 times today
write_through_audit
- post-run — playbook_memory has 1485 entries (ran 5 events, expected ≥ 3 new entries from this run)
Workers touched across the week
8 distinct workers made it through to a decision. Every one is accounted for below — no-shows flagged, rebookings noted, everyone visible.
| Worker ID | Name | Events | Outcome |
|---|---|---|---|
| W500K-18400 | Amy S. Jenkins | 10:30 baseline_fill | booked |
| W500K-6479 | Christina Watson | 10:30 baseline_fill | booked |
| W500K-21258 | Katherine Q. Gutierrez | 14:30 emergency | booked |
| W500K-38390 | Gary O. Stewart | 14:30 emergency | booked |
| W500K-26653 | Dorothy D. Lewis | 14:30 emergency | booked |
| W500K-41683 | Lauren F. Moore | 14:30 emergency | booked |
| 33537 | Susan X. Ruiz | 12:00 misplacement | booked |
| undefined | Amy S. Jenkins | 10:30 | no_show |
Discovered patterns (meta-index)
What the system identified across semantically-similar past fills as each event ran:
- 10:30 baseline_fill (Machine Operator): Across 25 similar past playbooks (13 workers examined) · recurring certifications: OSHA-10 (69%) · archetype mostly: communicator · reliability median 0.89 (range 0.59–1.00)
- 10:00 recurring (Receiving Clerk): —
- 09:30 expansion (Packer): —
- 14:30 emergency (Assembler): Across 25 similar past playbooks (14 workers examined) · recurring certifications: OSHA-10 (64%) · archetype mostly: communicator · reliability median 0.89 (range 0.59–1.00)
- 12:00 misplacement (Warehouse Associate): Across 25 similar past playbooks (13 workers examined) · recurring certifications: OSHA-10 (69%) · archetype mostly: communicator · reliability median 0.89 (range 0.59–1.00)
Narrative
- 3/5 events reached consensus.
- Final roster: 7 bookings across 1 distinct workers.
- Workers touched (booked, failed, or otherwise decided): 8.
- Playbook citations across the day: 0 (proof the feedback loop fired across events).
- Dropped events: 10:00 recurring, 09:30 expansion.