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>
6 lines
831 B
Markdown
6 lines
831 B
Markdown
# Cross-day lesson — Ironclad Industrial, 2026-04-22
|
|
|
|
_Generated by `gpt-oss:120b` in 2.7s. Based on 5 events + 2 mid-day checkpoints._
|
|
|
|
Before dispatching any role, always query the current pool status and verify that the pool value is known; if it is unknown, pause and request an updated pool snapshot rather than proceeding with a fill attempt. After each successful fill, immediately update the internal turn counter and recalculate drift thresholds to keep them in sync with the latest pool size. Ensure every JSON block (plan, tool_call, etc.) is fully closed and separated before moving to the next step, as an unclosed object will halt the entire workflow. Finally, run a quick sanity check after any emergency or misplacement event to confirm that drift flags have been reset before continuing with regular expansions.
|