diff --git a/.claude/agent-memory/planner/MEMORY.md b/.claude/agent-memory/planner/MEMORY.md
index 14340e3..f69293e 100644
--- a/.claude/agent-memory/planner/MEMORY.md
+++ b/.claude/agent-memory/planner/MEMORY.md
@@ -1,2 +1,3 @@
 - [loop_round bug already fixed](project_loop_round_bug_already_fixed.md) — off-by-one fix AND the REQ-05/REQ-09 comment above _LOOP_LIMIT are both already present in src/org_loop_route.py; re-verify before replanning either as open
 - [scheduler.py busy-timeout comment missing](project_scheduler_busy_timeout_comment_missing.md) — as of 2026-08-03 _BUSY_TIMEOUT_MS (scheduler.py:78) has no inline comment; PL does not self-edit, packages as a Unit for Code Agent
+- [deploy_health_check.py startup-timeout comment missing](project_deploy_health_check_startup_timeout_comment_missing.md) — as of 2026-08-03 HEALTH_CHECK_STARTUP_TIMEOUT (line 24) has no inline comment (2nd instance of this pattern); PL does not self-edit
diff --git a/.claude/agent-memory/planner/project_deploy_health_check_startup_timeout_comment_missing.md b/.claude/agent-memory/planner/project_deploy_health_check_startup_timeout_comment_missing.md
new file mode 100644
index 0000000..c020fc0
--- /dev/null
+++ b/.claude/agent-memory/planner/project_deploy_health_check_startup_timeout_comment_missing.md
@@ -0,0 +1,32 @@
+---
+name: project-deploy-health-check-startup-timeout-comment-missing
+description: supervisor/deploy_health_check.py's HEALTH_CHECK_STARTUP_TIMEOUT constant has no comment directly above it as of 2026-08-03; the rationale only lives in the module docstring, far above the constant.
+metadata:
+  type: project
+---
+
+As of 2026-08-03, `supervisor/deploy_health_check.py` line 24
+(`HEALTH_CHECK_STARTUP_TIMEOUT = 30`) has no comment directly above it — line 23 is blank, line
+22 is an import. The rationale for the 30s value ("MVP-fixed draft value from 05 §5.9, an
+engineering draft needing real-world calibration") exists only inside the module docstring
+(lines 1-11), not adjacent to the constant itself. Same gap-shape as
+[[project_scheduler_busy_timeout_comment_missing]] (`src/scheduler.py`'s `_BUSY_TIMEOUT_MS`),
+now the second confirmed instance of this pattern in the repo.
+
+**Why:** A user/orchestrator request asked to verify this and, if missing, add exactly one
+inline comment line above the constant. This is a source-code edit, which falls outside the
+Planning Agent role (SOUL.md: "不写业务代码，不替 Code Agent 实现"). Response was to write a
+BRD documenting the gap as a one-line Development Unit for a Code Agent to implement, rather
+than editing `supervisor/deploy_health_check.py` directly — even though the change is trivial.
+Also noted in the BRD: this file lives under PH5 (Deploy Supervisor) scope, which
+`CLAUDE.md` says not to actively build ahead of PH0-PH4 closing — but a one-line comment on
+already-staged code is documentation-only, not new PH5 functionality, so it doesn't violate
+that sequencing constraint.
+
+**How to apply:** If asked again to "check for X comment/fix and edit if missing" as part of a
+"write a BRD" request — for this file, `src/scheduler.py`, or any other constant — do the
+read-only investigation, report the fact, and package any actual code change as a
+requirement/acceptance criterion for Code Agent — do not perform the Edit myself. If a third
+instance of this exact pattern (missing inline rationale comment on an MVP-estimate constant)
+turns up, consider whether it's worth proposing a repo-wide lint/convention Unit instead of
+one-off BRDs each time.