docs(spec): correct the CLI spec's stale status on main - #21
Conversation
I marked it shipped on its branch after #18 had already merged, so main kept the in-progress copy — the record disagreeing with the work, which is the one thing a spec must not do. Outcome rewritten too: it still described waiting for a merge that had already happened, under the status rule that has since been corrected. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
There was a problem hiding this comment.
Approving — a record correction
Docs only. main's copy of the CLI spec claimed in-progress for work merged as 2fafb9f, because the shipped marking landed on the branch after the merge and never travelled. The Outcome paragraph carried the same staleness, describing a wait for a merge that had already happened.
Verified against main: the spec is 12/12 with every Validation line answered, and #18 is MERGED. So shipped is the accurate value, not an optimistic one.
Worth noting the general failure rather than just this instance: a spec marked shipped on its branch after its PR merges can never reach main. Marking it before opening the PR — which is what the status definition now implies, since shipped means "ready for a human to review" — avoids the whole class.
npm run check green. No source, no tests.
Approved.
specs/2026-08-02-cli-open-and-backup/plan.mdonmainstill says in-progress for work that merged as2fafb9f.Cause: I marked it
shippedon its own branch after #18 had already been merged, so the correction never reachedmain. The Outcome paragraph was stale in the same way — it described waiting for a merge that had already happened, under the status rule that has since been corrected inspecs/README.md.A record that disagrees with the work is the one thing a spec must not do, so this is a correction rather than a tidy-up.
Docs only — no source, no test.
npm run checkgreen.🤖 Generated with Claude Code