41b0a99ed2
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>
49 lines
2.1 KiB
JSON
49 lines
2.1 KiB
JSON
[
|
||
{
|
||
"date": "2026-04-30",
|
||
"client": "Parallel Machining",
|
||
"cities": "Flint",
|
||
"states": "MI",
|
||
"events_total": 3,
|
||
"events_ok": 0,
|
||
"checkpoint_count": 1,
|
||
"model": "gpt-oss:20b",
|
||
"cloud": false,
|
||
"lesson": "** \nBefore any shift, verify that all tools are calibrated and that the pool assignment is set—an unset pool causes immediate failure. For emergency tasks, perform a quick drift check and recalibrate tools on the spot to mitigate drift risk. Apply the same pre‑check routine to baseline and recurring tasks to avoid cascading failures. Keep a ready‑to‑run calibration script in the agent’s routine for rapid deployment.",
|
||
"checkpoints": [
|
||
{
|
||
"after": "12:30",
|
||
"risk": "Drift risk: emergency Packer Flint",
|
||
"hint": "Address drift flags: calibrate tools,"
|
||
}
|
||
],
|
||
"created_at": "2026-04-21T01:56:32.212Z",
|
||
"file": "2026-04-30_Parallel_Machining_1776736592212.json"
|
||
},
|
||
{
|
||
"date": "2026-04-27",
|
||
"client": "Pioneer Assembly",
|
||
"cities": "Flint",
|
||
"states": "MI",
|
||
"events_total": 5,
|
||
"events_ok": 0,
|
||
"checkpoint_count": 2,
|
||
"model": "gpt-oss:20b",
|
||
"cloud": false,
|
||
"lesson": "** \nBefore launching any shift, run a quick schema validation on the executor’s output to ensure the JSON array is properly closed and matches the planner’s expected format. Pre‑fetch and cache the candidate pool for each role so the planner has available candidates (avoiding zero turns). Finally, verify tool calibration—especially for roles like Shipping Clerk—in the target location before deployment to prevent drift.",
|
||
"checkpoints": [
|
||
{
|
||
"after": "14:00",
|
||
"risk": "JSON parse failure risk",
|
||
"hint": "Check executor output formatting; ensure JSON array closed properly; validate before sending to planner."
|
||
},
|
||
{
|
||
"after": "17:00",
|
||
"risk": "Shipping Clerk drift risk",
|
||
"hint": "Verify Shipping Clerk tool calibration in Flint before deployment to avoid drift."
|
||
}
|
||
],
|
||
"created_at": "2026-04-21T01:50:40.482Z",
|
||
"file": "2026-04-27_Pioneer_Assembly_1776736240482.json"
|
||
}
|
||
] |