docs(skills): add spec-start-github for issue-driven spec PRs - #821
Conversation
Add a thin skill that binds spec-start to GitHub (issue as canonical prompt, agent/<issue>-spec-* branches, local gh pr create vs harness post-script). Cross-link from spec-start for pairing. Closes fullsend-ai#803 Co-authored-by: Cursor <cursoragent@cursor.com>
|
fullsend review is working on this — view logs |
Site previewPreview: https://6982a52a-site.fullsend-ai.workers.dev Commit: |
ReviewFindingsMedium
Low
Previous runReview: #821Head SHA: f441e1b SummaryClean documentation-only PR that adds the FindingsNo findings. FooterOutcome: approve |
![fullsend-ai-review[bot]](https://avatars.githubusercontent.com/in/3478501?s=60&v=4){kind=small}
ralphbean
left a comment
ralphbean
left a comment
There was a problem hiding this comment.
LGTM. Two non-blocking notes inline.
Resolve spec-start description conflict: keep main's qna.md format note and the PR's spec-start-github pairing. Lead spec-start-github description with "Use when..." per review. Use truncated test PEM stub to satisfy gitleaks. Co-authored-by: Cursor <cursoragent@cursor.com>
|
Babysit update: merged `main` (resolved `spec-start` description conflict — kept the `qna.md` format sentence and the `spec-start-github` pairing). Updated `spec-start-github` frontmatter to lead with Use when... per review. CI is green and the branch is mergeable. |
fullsend-ai-review
Bot
added
the
requires-manual-review
3 participants
Summary
skills/spec-start-github/SKILL.md: thin GitHub layer (issue fetch, canonical prompt,agent/<issue>-spec-*branch rules, commit + localgh pr createvs harness handoff) that defers the headless spec procedure tospec-start(no duplicated checklist/templates).skills/spec-start/SKILL.mdwith cross-links so the two skills are paired for issue-driven runs.Context
Implements fullsend-ai/fullsend#803 (sub-issue of planning-oriented agents / spec workflows).
Verification
make lint(staged skill files) passed locally.Made with Cursor