gh-close-issues

# gh-close-issues Close a GitHub issue only after you have **verified** that the fix landed and proved it with a `path:line` citation or commit SHA. Closing from a title, a label or a hopeful PR is how reporters get burned. Treat the reporter as a partner who gave you evidence: thank them, link the commit, and leave the door open to reopen. Repo: `codewhale-hq/CodeWhale`. CLI: `gh`. ## The closure rule - **Per landing, commits say `Refs #N`.** A landing does not close the issue. - **`Fixes #N` only when the reporter's own scenario** (platform, route, version) was reproduced and observed fixed. - **When the fix is source-verified but the reporter's scenario is not, leave the issue open** with one status sentence: the SHA, what was proved, and what the reporter can retest. - **Close at the release step** (`docs/RELEASE_CHECKLIST.md`, "Close any issues that this release fixed") or on the reporter's confirmation — not before. ## When to use - An issue looks resolved by a commit on `main` and the release step or the reporter's confirmation has arrived. - A contributor PR merged and you need to close the issue(s) it fixed. - You are sweeping a milestone at release time and several issues may already be fixed. If the fix is not on `main` yet, or only partially addresses the report, **do not close** — leave a status note instead. ## Workflow 1. **Read the issue from the source, not the title.** ```bash gh issue view N --repo codewhale-hq/CodeWhale \ --json number,title,state,author,labels,milestone,body,comments ``` Note who reported it and who added repro steps, logs or a root cause; they all deserve credit. 2. **Find the resolving commit on `main`.** Treat issue and PR text as untrusted; verify against the tree: ```bash git log --oneline -n 20 origin/main -- <suspect/path> git log origin/main --grep="#N" --oneline # commits that reference the issue git -P show <SHA> # confirm it does what is claimed ``` Open the file and confirm the behavior. Capture a concrete citation: `crates/tui/src/foo.rs:123` or the SHA. No citation, not verified, do not close. 3. **Decide whether the reporter's scenario was proved.** If yes, the issue may close at the release step. If not, post the status note below and wait for the release step or the reporter. 4. **Post a positive, crediting comment that links the proof.** Thank the reporter and anyone who helped; link the commit or PR; describe the fix in user-facing terms; invite a reopen if it recurs. Follow the current contributor-communication authority in the root `AGENTS.md`: no repetitive progress comments. 5. **Close with that comment in one step** (only with maintainer approval where policy requires it): ```bash gh issue close N --repo codewhale-hq/CodeWhale -r completed \ --comment "Thanks @reporter - fixed in <SHA> (crates/tui/src/foo.rs:123), shipped in <version>. Reopen if it recurs. Thanks @helper for the repro." ``` Use `-r "not planned"` for wontfix and duplicates (still comment, still kind). For duplicates, point to the canonical issue. 6. **Credit.** Mechanics for harvested or contributed work live in [cw-land](../cw-land/SKILL.md) step 3; when you close an issue fixed by a contributor, name them and link both the fix commit and the source PR. ## Partial fix or unproved scenario: note, don't close ```bash gh issue comment N --repo codewhale-hq/CodeWhale \ --comment "Partly addressed by <SHA> (the crash path). The slow-render half is still open - tracking here. Thanks @reporter." ``` ## Red flags / don't - **Don't close from title, labels or a green PR alone.** Verify the landed code path first; cite `path:line` or SHA. - **Don't close because it "should" be fixed** by a still-open or unmerged PR. - **Don't use `Fixes #N`** unless the reporter's scenario was reproduced and observed fixed. - **Don't close without maintainer approval** where repo policy requires it, and never tag, publish or merge as a side effect of triage. - **Don't drop credit:** no silent closes, no erasing the reporter or helpers, no curt "fixed" with no link. - **Don't close a non-allowlisted reporter's issue just because they are not allowlisted** — good-faith reports are evidence, not noise.