You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 442147d
Browse filesBrowse the repository at this point in the historyBrowse files
Copy file name to clipboardExpand all lines: AGENTS.md
+2Lines changed: 2 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -44,6 +44,8 @@ Every reproduced race requires a failing-before and passing-after regression. As
44
44
- Run final validation against the exact pushed head after the last change.
45
45
- Report current test counts separately from historical milestone counts.
46
46
- Before requesting automated review, report the current head, CI state, mergeability, unresolved threads, and deferred follow-up issues.
47
+
- In evidence records, label cited commits as baselines, intermediate checkpoints, or validated heads. Keep final exact-head results in a place that can name the resulting commit, such as the PR body or CI record.
47
48
- Reproduce summary-only review concerns or turn them into a concrete follow-up issue. Do not repeatedly patch vague wording without a failure case.
49
+
- A validation fixture for a summary-only concern must assert the disputed intermediate representation or state before using a downstream outcome as evidence that the concern was exercised.
48
50
- For each review round, record what changed, what was declined and why, and the regression evidence. Re-request review until a round returns no new findings.
49
51
- Treat review-lesson extraction as a merge gate. Before invoking merge, classify every review finding in the PR body as: covered by an existing rule (cite it), captured by a new rule in this branch (cite it), or one-off (record why). Do not merge until this audit is complete and every required `AGENTS.md` update is included in the reviewed head. Omit rules that merely repeat existing guidance.
Validated on 2026-09-23 before selecting real issues for the plan-indexed review experiment. The validation branch started from baseline `5a2685c`; final exact-head results are recorded in PR #17. These were summary-only concerns from PR #9, not reproduced defects. No production behavior changed during this validation.
4
+
5
+
## Renamed-file reassignment
6
+
7
+
A real-Git fixture reduces the demo plan to P1, declares `retry.ts` → `renamed.ts`, then creates an unplanned rename with a content edit. The review exposes removed and added text segments with `-` and `+` operations. The fixture resolves the removed side through `oldPath` and the added side through `path`, verifies both declared names, then manually assigns both segments. Each remains in scope.
8
+
9
+
Result: no defect reproduced. `ReviewService` evaluates a manual assignment against both `path` and `renamed_from`. The integration fixture remains as coverage.
10
+
11
+
## Literal pathspec metacharacters
12
+
13
+
A planting fixture uses the literal declared filename `*.txt` beside a decoy `a.txt`. An owned commit edits the literal file; a later commit removes it while retaining the decoy. Direct controls prove `git ls-tree ... -- '*.txt'` returns no match and that `ls-tree` rejects `:(glob)` pathspec magic as unsupported. The plant helper then rejects the history at its declared-file transition check with `Declared plant needs a regular file retained through the remaining history.`
14
+
15
+
Result: no defect reproduced. The decoy does not satisfy the literal transition check. The fixture remains as coverage for Git pathspec metacharacters.
16
+
17
+
## Accepted-change status styling
18
+
19
+
The existing browser flow accepts an unplanned change, opens the Accepted row, and reloads it. The status reads `Accepted outside plan` and uses the existing muted treatment. A proposed error-color assertion failed because the rendered class is `muted`, confirming the summary did not describe the current styling accurately enough to imply a patch.
20
+
21
+
Result: no defect reproduced. The summary supplied no expected color, state token, contrast failure, or browser failure. The explicit label already distinguishes the acknowledged exception, so no new color semantics were inferred and the existing browser coverage remains unchanged.
22
+
23
+
## Gate result
24
+
25
+
All three concerns are resolved as validated behavior. No issue supplied a failing case that justified a production patch. Issue #3's manual assignment and paired human go/no-go experiment remain pending and must use the frozen protocol in `review-protocol.md`.
it('persists bounded per-item notes without creating a plan revision',()=>{
20
38
const{service,config}=fixture();constview=service.load();service.act({action:'note',item:'P1',kind:'question',text:'Why this limit?',token:view.token});
21
39
constreopened=newReviewService(config);services.push(reopened);expect(reopened.load().notes[0]!.text).toBe('Why this limit?');expect(reopened.load().plan.revision).toBe(1);
0 commit comments