Commit c425ab85a5059139602a00f0b1e036890b518b04
graphlemon: merge 044315ed-c80e-4866-a582-92efda9e9315 (retried after cleanup)
Showing
1 changed file
with
8 additions
and
0 deletions
.claude/agent-memory/planner/project_loop_round_bug_already_fixed.md
| ... | ... | @@ -32,3 +32,11 @@ still open — it may have been fixed between conversations, or the request may |
| 32 | 32 | the *planning process* (e.g. via the self-hosting E2E harness) rather than asking for a real code |
| 33 | 33 | change. Surface "already fixed, has regression coverage" as a first-class BRD finding rather than |
| 34 | 34 | silently manufacturing a fix for a non-existent defect. |
| 35 | + | |
| 36 | +**Update 2026-08-03:** the documentation follow-up ("add a one-line comment above `_LOOP_LIMIT = 3` | |
| 37 | +explaining the REQ-05/REQ-09 loop_round boundary") is also already done. As of this date, | |
| 38 | +`src/org_loop_route.py:38` reads exactly `# REQ-05/REQ-09 (docs/sdd/06-验收与测试锚点.md) | |
| 39 | +loop_round boundary: loop_round >= _LOOP_LIMIT forces task.status -> blocked_pending_human.` | |
| 40 | +directly above line 39's `_LOOP_LIMIT = 3`. If asked again to add this comment, re-verify the file | |
| 41 | +first instead of trusting "no comment there" in the goal description, and do not duplicate/re-add | |
| 42 | +a comment that already exists. | ... | ... |