chore: add real content that was sitting untracked

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>
2026-05-02 22:22:10 -05:00
parent 6e34ef7baf
commit 41b0a99ed2
788 changed files with 107142 additions and 0 deletions
@@ -0,0 +1,2 @@
+{"after_event":"11:45","event_kind":"emergency","ok":true,"model":"gpt-oss:120b","duration_secs":7.582,"risk":"Potential drift flags for upcoming role","hint":"Perform tool and pool sanity check before dispatch; ensure drift thresholds reset."}
+{"after_event":"15:30","event_kind":"misplacement","ok":true,"model":"gpt-oss:120b","duration_secs":3.655,"risk":"Invalid JSON response halts workflow","hint":"Validate and correctly close all JSON objects (plan, tool_call) before output; ensure proper separation of sections to avoid parse errors."}
@@ -0,0 +1 @@
+{"at":"14:00","kind":"expansion","operation":"fill: Tool & Die Maker x4 in Grand Rapids, MI","fills":[{"candidate_id":"W500K-49117","name":"Deborah D. Wilson"},{"candidate_id":"W500K-4243","name":"Jerry Thompson"},{"candidate_id":"W500K-14314","name":"Carolyn Lopez"},{"candidate_id":"W500K-15737","name":"Kyle H. Ramos"}],"turns":2,"duration_secs":12.711,"pool_size":133,"playbook_citations":[],"discovered_pattern":"Across 25 similar past playbooks (34 workers examined) · recurring certifications: OSHA-10 (47%) · archetype mostly: flexible · reliability median 0.77 (range 0.19–1.00)"}
@@ -0,0 +1 @@
+# Client emails — Ironclad Industrial, 2026-04-22
@@ -0,0 +1,5 @@
+# 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.
@@ -0,0 +1,70 @@
+[
+  {
+    "date": "2026-05-10",
+    "client": "Midway Distribution",
+    "cities": "Chicago",
+    "states": "IL",
+    "events_total": 3,
+    "events_ok": 2,
+    "checkpoint_count": 1,
+    "model": "gpt-oss:20b",
+    "cloud": false,
+    "lesson": "**  \nBefore assigning a Packer to Chicago, verify the candidate’s current assignment and cross‑check drift logs for recent misplacements. Pre‑calibrate packing tools and load a backup Packer into the pool to avoid a single point of failure. If a drift risk is flagged, immediately relocate the Packer or switch to a verified alternate. Keep the pool count updated and ensure it matches the required number of roles before each event.",
+    "checkpoints": [
+      {
+        "after": "09:30",
+        "risk": "Packer drift risk persists",
+        "hint": "Verify Packer's Chicago assignment, review drift logs, recalibrate tools, and consider temporary relocation."
+      }
+    ],
+    "created_at": "2026-04-21T02:14:12.176Z",
+    "file": "2026-05-10_Midway_Distribution_1776737652176.json"
+  },
+  {
+    "date": "2026-05-08",
+    "client": "Summit Industrial",
+    "cities": "Detroit",
+    "states": "MI",
+    "events_total": 2,
+    "events_ok": 2,
+    "checkpoint_count": 1,
+    "model": "gpt-oss:20b",
+    "cloud": false,
+    "lesson": "**  \nWhen a recurring shift precedes a baseline‑fill, pre‑fetch the candidate’s calendar for both slots to avoid double‑booking. Always populate the `f.reason` field for every assignment to prevent artifact errors. In this run, Sarah D. Cook was at risk; check her calendar before assigning her to the 13:00 shift. Keep pool counts and turns updated after each fill to maintain accurate metrics.",
+    "checkpoints": [
+      {
+        "after": "10:30",
+        "risk": "Double-booking risk for Sarah",
+        "hint": "Check Sarah D. Cook's calendar before assigning to the next shift; ensure f.reason is defined to avoid artifact errors."
+      }
+    ],
+    "created_at": "2026-04-21T02:10:51.019Z",
+    "file": "2026-05-08_Summit_Industrial_1776737451019.json"
+  },
+  {
+    "date": "2026-05-05",
+    "client": "Midway Distribution",
+    "cities": "Indianapolis",
+    "states": "IN",
+    "events_total": 4,
+    "events_ok": 3,
+    "checkpoint_count": 2,
+    "model": "gpt-oss:20b",
+    "cloud": false,
+    "lesson": "**  \nBefore running an expansion, pre‑fetch local Packer data (or fallback to the nearest city) and validate that the executor returns well‑formed JSON; this prevents the “Missing local Packer data” failure.  \nWhen an emergency request arrives, immediately cross‑check the 16:00 schedule to ensure no worker is assigned twice, and correct any artifact errors in the scheduling logic that could cause double‑booking.  \nImplement automated checks that flag duplicate assignments and missing data before the agent commits to a fill.",
+    "checkpoints": [
+      {
+        "after": "10:30",
+        "risk": "Missing local Packer data",
+        "hint": "Add Indianapolis Packer records to the index or fallback to nearest city; verify executor returns valid JSON before execution."
+      },
+      {
+        "after": "12:00",
+        "risk": "16:00 double-booking risk",
+        "hint": "Verify no worker is assigned twice for 16:00; fix artifact error in scheduling logic."
+      }
+    ],
+    "created_at": "2026-04-21T02:04:50.585Z",
+    "file": "2026-05-05_Midway_Distribution_1776737090585.json"
+  }
+]
@@ -0,0 +1,74 @@
+# Scenario retrospective — Ironclad Industrial, 2026-04-22
+
+Executor: `qwen3.5:latest`   Reviewer: `qwen3:latest`   Draft: `qwen2.5:latest`   Overview(T3): `gpt-oss:120b (cloud)`
+Prior lessons loaded into executor context: **3** (from 2026-05-10, 2026-05-08, 2026-05-05)
+
+## Events
+
+| At | Kind | Role / Count | Pool | Fills | Turns | Dur(s) | Cites | Gaps |
+|---|---|---|---|---|---|---|---|---|
+| 07:00 | baseline_fill | Electrician × 5 | - | ✗ 0 | 0 | 24.9 | 0 | 1 |
+| 09:30 | expansion | Safety Coordinator × 8 | - | ✗ 0 | 0 | 48.7 | 0 | 1 |
+| 11:45 | emergency | Welder × 3 | - | ✗ 0 | 0 | 20.1 | 0 | 1 |
+| 14:00 | expansion | Tool & Die Maker × 4 | 133 | ✓ 4 | 2 | 12.7 | 0 | 4 |
+| 15:30 | misplacement | Electrician × 1 | - | ✗ 0 | 0 | 26.8 | 0 | 1 |
+
+## Final roster
+
+| Worker | Booked | Role | City, ST | Status |
+|---|---|---|---|---|
+| undefined Deborah D. Wilson | 14:00 | Tool & Die Maker | Grand Rapids, MI | confirmed |
+| undefined Jerry Thompson | 14:00 | Tool & Die Maker | Grand Rapids, MI | confirmed |
+| undefined Carolyn Lopez | 14:00 | Tool & Die Maker | Grand Rapids, MI | confirmed |
+| undefined Kyle H. Ramos | 14:00 | Tool & Die Maker | Grand Rapids, MI | confirmed |
+
+## Gap signals
+
+### drift_or_tool
+- **07:00** — aborted — 3 consecutive drift flags
+- **09:30** — aborted — 3 consecutive drift flags
+- **11:45** — aborted — 3 consecutive drift flags
+- **15:30** — invalid JSON from executor: JSON Parse error: Expected '}' | raw: {"kind":"plan","steps":["Expand search to nearby cities (South Bend, IN) and adjacent states (OH, MI) for Electricians with availability > 0.5.","Verify top candidates via SQL to confirm role, location, and availability.","Select the best verified candidate and propose_done."]{"kind":"tool_call","to
+
+### double_book
+- **14:00** — undefined Jerry Thompson already booked for 14:00
+- **14:00** — undefined Carolyn Lopez already booked for 14:00
+- **14:00** — undefined Kyle H. Ramos already booked for 14:00
+
+### artifact
+- **14:00** — undefined is not an object (evaluating 'f.reason.slice')
+
+### fairness
+- _cross-event_ — Deborah D. Wilson (undefined) booked 4 times today
+
+### write_through_audit
+- _post-run_ — playbook_memory has 1566 entries (ran 5 events, expected ≥ 1 new entries from this run)
+
+## Workers touched across the week
+
+4 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-49117 | Deborah D. Wilson | 14:00 expansion | booked |
+| W500K-4243 | Jerry Thompson | 14:00 expansion | booked |
+| W500K-14314 | Carolyn Lopez | 14:00 expansion | booked |
+| W500K-15737 | Kyle H. Ramos | 14:00 expansion | booked |
+
+## Discovered patterns (meta-index)
+
+What the system identified across semantically-similar past fills as each event ran:
+
+- **07:00 baseline_fill** (Electrician): —
+- **09:30 expansion** (Safety Coordinator): —
+- **11:45 emergency** (Welder): —
+- **14:00 expansion** (Tool & Die Maker): Across 25 similar past playbooks (34 workers examined) · recurring certifications: OSHA-10 (47%) · archetype mostly: flexible · reliability median 0.77 (range 0.19–1.00)
+- **15:30 misplacement** (Electrician): —
+
+## Narrative
+
+- 1/5 events reached consensus.
+- Final roster: 4 bookings across 1 distinct workers.
+- Workers touched (booked, failed, or otherwise decided): 4.
+- Playbook citations across the day: 0 (proof the feedback loop fired across events).
+- Dropped events: 07:00 baseline_fill, 09:30 expansion, 11:45 emergency, 15:30 misplacement.
@@ -0,0 +1,128 @@
+[
+  {
+    "event": {
+      "kind": "baseline_fill",
+      "at": "07:00",
+      "role": "Electrician",
+      "count": 5,
+      "city": "Gary",
+      "state": "IN",
+      "shift_start": "07:00 AM",
+      "scenario_note": "Gary IN has ZERO Electricians in the index. Local WILL fail this. Cloud should diagnose no-supply and recommend pivoting to Chicago IL (40min drive) or relaxing to 'Maintenance Tech'."
+    },
+    "ok": false,
+    "fills": [],
+    "turns": 0,
+    "duration_secs": 24.921,
+    "error": "aborted — 3 consecutive drift flags",
+    "gap_signals": [
+      "drift_or_tool: aborted — 3 consecutive drift flags"
+    ]
+  },
+  {
+    "event": {
+      "kind": "expansion",
+      "at": "09:30",
+      "role": "Safety Coordinator",
+      "count": 8,
+      "city": "Peoria",
+      "state": "IL",
+      "shift_start": "09:30 AM",
+      "scenario_note": "Safety Coordinator is the rarest role overall (~4500 nationally). 8× in a mid-sized city with availability > 0.5 is genuinely tight. Cloud should either confirm or suggest multi-city sourcing."
+    },
+    "ok": false,
+    "fills": [],
+    "turns": 0,
+    "duration_secs": 48.719,
+    "error": "aborted — 3 consecutive drift flags",
+    "gap_signals": [
+      "drift_or_tool: aborted — 3 consecutive drift flags"
+    ]
+  },
+  {
+    "event": {
+      "kind": "emergency",
+      "at": "11:45",
+      "role": "Welder",
+      "count": 3,
+      "city": "Flint",
+      "state": "MI",
+      "shift_start": "12:00 PM",
+      "deadline": "13:30",
+      "scenario_note": "Flint MI has ZERO workers indexed — total data desert. Cloud must flag 'impossible supply' and recommend pivot (Detroit 60mi, Saginaw 40mi)."
+    },
+    "ok": false,
+    "fills": [],
+    "turns": 0,
+    "duration_secs": 20.138,
+    "error": "aborted — 3 consecutive drift flags",
+    "gap_signals": [
+      "drift_or_tool: aborted — 3 consecutive drift flags"
+    ]
+  },
+  {
+    "event": {
+      "kind": "expansion",
+      "at": "14:00",
+      "role": "Tool & Die Maker",
+      "count": 4,
+      "city": "Grand Rapids",
+      "state": "MI",
+      "shift_start": "14:00 PM",
+      "scenario_note": "Tool & Die Maker is scarce (~9000 total). 4× in Grand Rapids, availability > 0.5 AND reliability > 0.75. Tight but solvable if playbook_memory has history; cloud should prioritize proven performers."
+    },
+    "ok": true,
+    "fills": [
+      {
+        "candidate_id": "W500K-49117",
+        "name": "Deborah D. Wilson"
+      },
+      {
+        "candidate_id": "W500K-4243",
+        "name": "Jerry Thompson"
+      },
+      {
+        "candidate_id": "W500K-14314",
+        "name": "Carolyn Lopez"
+      },
+      {
+        "candidate_id": "W500K-15737",
+        "name": "Kyle H. Ramos"
+      }
+    ],
+    "turns": 2,
+    "duration_secs": 12.711,
+    "gap_signals": [
+      "double_book: undefined Jerry Thompson already booked for 14:00",
+      "double_book: undefined Carolyn Lopez already booked for 14:00",
+      "double_book: undefined Kyle H. Ramos already booked for 14:00",
+      "artifact: undefined is not an object (evaluating 'f.reason.slice')"
+    ],
+    "sources_first_score": 0.72152853,
+    "sources_last_score": 0.51832277,
+    "pool_size": 133,
+    "playbook_citations": [],
+    "discovered_pattern": "Across 25 similar past playbooks (34 workers examined) · recurring certifications: OSHA-10 (47%) · archetype mostly: flexible · reliability median 0.77 (range 0.19–1.00)"
+  },
+  {
+    "event": {
+      "kind": "misplacement",
+      "at": "15:30",
+      "role": "Electrician",
+      "count": 1,
+      "city": "Gary",
+      "state": "IN",
+      "shift_start": "15:30 PM",
+      "replaces_event": "07:00",
+      "scenario_note": "Refilling 1× Electrician in Gary after a no-show. Same data desert as event 1 — cloud should recognize the repeat and recommend the SAME pivot it gave earlier, proving it learns within-run."
+    },
+    "ok": false,
+    "fills": [],
+    "turns": 0,
+    "duration_secs": 26.796,
+    "error": "invalid JSON from executor: JSON Parse error: Expected '}' | raw: {\"kind\":\"plan\",\"steps\":[\"Expand search to nearby cities (South Bend, IN) and adjacent states (OH, MI) for Electricians with availability > 0.5.\",\"Verify top candidates via SQL to confirm role, location, and availability.\",\"Select the best verified candidate and propose_done.\"]{\"kind\":\"tool_call\",\"to",
+    "gap_signals": [
+      "drift_or_tool: invalid JSON from executor: JSON Parse error: Expected '}' | raw: {\"kind\":\"plan\",\"steps\":[\"Expand search to nearby cities (South Bend, IN) and adjacent states (OH, MI) for Electricians with availability > 0.5.\",\"Verify top candidates via SQL to confirm role, location, and availability.\",\"Select the best verified candidate and propose_done.\"]{\"kind\":\"tool_call\",\"to"
+    ]
+  }
+]
@@ -0,0 +1,34 @@
+[
+  {
+    "name": "Deborah D. Wilson",
+    "booked_for": "14:00",
+    "role": "Tool & Die Maker",
+    "city": "Grand Rapids",
+    "state": "MI",
+    "status": "confirmed"
+  },
+  {
+    "name": "Jerry Thompson",
+    "booked_for": "14:00",
+    "role": "Tool & Die Maker",
+    "city": "Grand Rapids",
+    "state": "MI",
+    "status": "confirmed"
+  },
+  {
+    "name": "Carolyn Lopez",
+    "booked_for": "14:00",
+    "role": "Tool & Die Maker",
+    "city": "Grand Rapids",
+    "state": "MI",
+    "status": "confirmed"
+  },
+  {
+    "name": "Kyle H. Ramos",
+    "booked_for": "14:00",
+    "role": "Tool & Die Maker",
+    "city": "Grand Rapids",
+    "state": "MI",
+    "status": "confirmed"
+  }
+]
@@ -0,0 +1 @@
+# SMS drafts — Ironclad Industrial, 2026-04-22
				`@@ -0,0 +1 @@`
				{"at":"14:00","kind":"expansion","operation":"fill: Tool & Die Maker x4 in Grand Rapids, MI","fills":[{"candidate_id":"W500K-49117","name":"Deborah D. Wilson"},{"candidate_id":"W500K-4243","name":"Jerry Thompson"},{"candidate_id":"W500K-14314","name":"Carolyn Lopez"},{"candidate_id":"W500K-15737","name":"Kyle H. Ramos"}],"turns":2,"duration_secs":12.711,"pool_size":133,"playbook_citations":[],"discovered_pattern":"Across 25 similar past playbooks (34 workers examined) · recurring certifications: OSHA-10 (47%) · archetype mostly: flexible · reliability median 0.77 (range 0.19–1.00)"}
				`@@ -0,0 +1 @@`
				`# Client emails — Ironclad Industrial, 2026-04-22`
				`@@ -0,0 +1 @@`
				`# SMS drafts — Ironclad Industrial, 2026-04-22`