Commit 69567be4c518f784b4dc6fa31b40b120498ed59c

Authored by graphlemon
2 parents 93b76bdc 761681ae

graphlemon: merge 4377f28d-6610-4891-a1de-58745b01f55d

.claude/agent-memory/coder/MEMORY.md
1 - [Loop-limit comment task keeps re-dispatching with zero diff](task_dispatch_loop_limit_comment_repeat.md) — src/org_loop_route.py:38 comment already exists; verify and report, don't re-add 1 - [Loop-limit comment task keeps re-dispatching with zero diff](task_dispatch_loop_limit_comment_repeat.md) — src/org_loop_route.py:38 comment already exists; verify and report, don't re-add
  2 +- [deploy_health_check timeout comment fixed 2026-08-03](project_deploy_health_check_startup_timeout_comment_resolved.md) — PH5 already shipped, old "defer" reasoning is stale; check dev-plan closure before citing phase gates
.claude/agent-memory/coder/project_deploy_health_check_startup_timeout_comment_resolved.md 0 → 100644
  1 +---
  2 +name: project-deploy-health-check-startup-timeout-comment-resolved
  3 +description: HEALTH_CHECK_STARTUP_TIMEOUT inline comment gap in supervisor/deploy_health_check.py was fixed 2026-08-03 -- do not re-investigate or re-defer this as PH5-gated, PH5 already shipped
  4 +metadata:
  5 + type: project
  6 +---
  7 +
  8 +As of 2026-08-03, `supervisor/deploy_health_check.py` line ~24 (`HEALTH_CHECK_STARTUP_TIMEOUT = 30`)
  9 +now has a one-line comment directly above it: `# MVP engineering estimate per 05 §5.9 -- needs
  10 +real-world calibration, not an authoritative constant.` Diff was exactly +1 line, no other changes.
  11 +Verified with `python3 -m py_compile` and `python3 -m unittest supervisor.test_deploy_health_check`
  12 +(11/11 passed, unchanged).
  13 +
  14 +**Why this is different from the prior three "PH5-gated, decline to edit" recurrences**: those
  15 +Planner-role investigations (logged in the global Claude memory store as
  16 +`project_health_check_timeout_comment_ph5_deferred.md`, and in
  17 +`.claude/agent-memory/planner/project_deploy_health_check_startup_timeout_comment_missing.md`)
  18 +correctly declined based on `CLAUDE.md`'s "PH5 is not in scope yet" text — but that text describes
  19 +the repo's PH0 pre-scaffolding state and is stale. By the time this Coder task ran,
  20 +`docs/development-plan.md` showed PH5 (U-034~U-042) fully implemented and closed, including
  21 +`supervisor/deploy_health_check.py` itself (added in U-038, 2026-08-02) and a working self-hosting
  22 +e2e loop (U-042). The PH5-gating rationale no longer applied — the file is live production code
  23 +being iterated on normally, not premature work.
  24 +
  25 +**How to apply:** If this exact task resurfaces again, check current state first — the comment is
  26 +now present, so report that and make no changes (matches the "if yes, report and stop" branch of
  27 +the standard request pattern, see [[task_dispatch_loop_limit_comment_repeat]]). Don't reflexively
  28 +apply the old PH5-deferral reasoning without re-checking `docs/development-plan.md` for phase
  29 +closure status — `CLAUDE.md`'s narrative text can go stale relative to actual repo progress in a
  30 +fast-moving self-hosting project like this one; always cross-check against the development plan's
  31 +own "已收口" (closed) markers before citing a phase gate as a reason to decline work.
supervisor/deploy_health_check.py
@@ -21,6 +21,7 @@ import uuid @@ -21,6 +21,7 @@ import uuid
21 21
22 from supervisor import deploy_supervisor 22 from supervisor import deploy_supervisor
23 23
  24 +# MVP engineering estimate per 05 §5.9 -- needs real-world calibration, not an authoritative constant.
24 HEALTH_CHECK_STARTUP_TIMEOUT = 30 25 HEALTH_CHECK_STARTUP_TIMEOUT = 30
25 HEALTH_CHECK_HTTP_TIMEOUT = 10 26 HEALTH_CHECK_HTTP_TIMEOUT = 10
26 HEALTH_CHECK_CANARY_TIMEOUT = 120 27 HEALTH_CHECK_CANARY_TIMEOUT = 120