docs(living): mirror operating log (TODAY / restart checklist / claim-path)

[DealAppSeo]

Mirrors three internal operating-log docs into docs/living/ for GitHub parity with E:\dev\living-docs, per standing order.

• CLAIM_PATH.md — root-cause of the 8–83 day claim stall: every pending task was assigned_to an agent not in the running fleet, so the claim predicate excluded them. Claim path proven healthy (open-pool probe claimed in 11s). H1–H5 falsified.
• AGENT_RESTART_CHECKLIST.md — Sean-only restart runbook; leads with the finding that loops are advancing, so restart is likely the wrong lever.
• TODAY.md — day plan / OPEN state.

⚠️ Note: this is a public repo and these are internal operating docs (no secrets — scanned; project ref only, which canon treats as non-secret). Flagging in case internal-ops content in the public builder repo isn't intended — merge is your call.

Merge stays with Sean.

[strix-security]

Strix Security Review

Warning

This pull request has 14 commits after the last Strix review (891a212). Strix has not reviewed these changes.
Automatic review on push is off for this repository. To review the latest changes, tag @strix-security in a comment, or turn on re-review on push.

1 open security finding on this PR:

• 🟢 LOW Internal operational infrastructure details committed to public repository — docs/living/CLAIM_PATH.md:2

Review summary

Reviewed all four changed files (README.md and the three new docs/living/ operating documents). The PR is documentation-only and contains no hardcoded secrets, credentials, API keys, tokens, or private keys (verified with gitleaks, trufflehog, and manual pattern search). The substantive security concern is an internal operational information disclosure: the new documents publish a separate internal system's production database identifier, PostgreSQL table/view/column names, internal agent service names, task identifiers, health-endpoint fields, and environment variable names to this public repository. This was independently validated and filed as "Internal operational infrastructure details committed to public repository". The README change and the remaining documentation content introduce no other security issues.

_{Fixed the findings? re-run the review, or tag @strix-security in a PR comment to run a fresh review.}

_{Updated for 891a212.}

---

Reviewed by Strix
_{Re-run review · Configure security review settings}

[strix-security]

Strix flagged a new security finding below. See the pinned summary comment for the full PR status.

[strix-security]

🟢 Internal operational infrastructure details committed to public repository

Severity: LOW · CWE-200

The pull request publishes three internal operating documents under docs/living/ that disclose production infrastructure identifiers to a public repository. The documents reveal an internal PostgreSQL database name, database schema/view/table and column names, twelve internal agent service names, internal task identifiers and types, health endpoint field names, and environment variable names. The disclosure does not include credentials, connection strings, private keys, or PII, so the exposure is limited to operational/architectural intelligence.

Prompt to fix with AI

This is a security vulnerability found during a code review.

Vulnerability: Internal operational infrastructure details committed to public repository
Severity: LOW
CWE: CWE-200

The pull request publishes three internal operating documents under `docs/living/` that disclose production infrastructure identifiers to a public repository. The documents reveal an internal PostgreSQL database name, database schema/view/table and column names, twelve internal agent service names, internal task identifiers and types, health endpoint field names, and environment variable names. The disclosure does not include credentials, connection strings, private keys, or PII, so the exposure is limited to operational/architectural intelligence.

Location: docs/living/CLAIM_PATH.md:2-2
Context: Production database identifier
```
**Diagnosed** 2026-09-10 (CC, standing order Loop 1). Live evidence against Trinity prod `qnnpjhlxljtqyigedwkb`.
```

Location: docs/living/CLAIM_PATH.md:24-32
Context: Internal task IDs/types and 12 internal agent service names
```
| id | task_type | assigned_to | claim_count |
|---|---|---|---|
| 435104 | remediation | `trinity-gemini-antigravity` | 0 |
| 435111 | e2e_loop | `trinity-grok-code` | 0 |
| 435096 | a2a_work | `trinity-gemini-antigravity` | 0 |
| 435097 | verification | `trinity-grok-code` | 0 |
| 435099 | tuning | `trinity-cowork-executor` | 0 |

Running fleet = apm, chesed, gcm, hdm, mel, nexus, orch, shofet, sophia, torch, veritas, w3c. For any of them as `$1`, a row with a non-NULL `assigned_to` that isn't their own name matches **neither** predicate branch → excluded. So the open pool is **empty for the running fleet**; the work is directed at external CLI agents (Gemini Antigravity, Grok Code, Cowork) that aren't running. `insert_source='cowork-cl-session'` corroborates: a Cowork CLI session created and hand-assigned them.
```

Location: docs/living/AGENT_RESTART_CHECKLIST.md:18-25
Context: Health endpoint fields and environment variable names
```
For each of the 12 `trinity-*` services, capture from the Railway dashboard / `GET /health`:
| Field | Where | Healthy looks like |
|---|---|---|
| `HEARTBEAT_MODE` | service env var | `throttled` or `off` — **never `full`** (full = ~8.6M writes/day storm) |
| `loopCount` | `/health` body | advancing between two reads 5 min apart |
| `lastIterationAt` | `/health` body | < 15 min old |
| last `llm_call_log` | DB `llm_call_log` for that agent | a row in the last hour if it's meant to be working |
| `RAILWAY_GIT_COMMIT_SHA` | service env / deploy | matches `main` HEAD; a stale SHA = running old code |
```

How to fix:
Remove the internal operating documents from the public repository, or redact the internal identifiers (database name, service names, schema/table/view and column names, health endpoint fields, and environment variable names) before publication. Keep day-to-day operational runbooks in the private internal documentation location rather than mirroring them to a public builder repository. Re-scan the redacted content with a secrets scanner prior to merge to confirm no credentials remain.

Please fix this vulnerability. If you propose a fix, make it concise and minimal.

_{React 👍 / 👎 to tune Strix for this repo. A repo collaborator (or the PR author) can resolve this thread to dismiss the finding.}

[strix-security]

Internal task IDs/types and 12 internal agent service names

[strix-security]

Health endpoint fields and environment variable names