Extract /deploy bump-verify-push sequence into bump-and-tag.py#168
Conversation
Review SummaryVerdict: APPROVE Findings
No correctness, security, or convention issues found. The skill markdown collapse is consistent across all four adapters and the source template, and the generated-file hashes were refreshed via sync. |
Review Summary (re-review of commit 45840a1)Verdict: APPROVE The follow-up commit correctly addresses the prior NOTE finding about the stray-space rendering bug. The source template at Findings
|
Review SummaryVerdict: APPROVE FindingsNo issues found. The trailing-separator convention for |
Extract the bump-verify-push sequence from
/deployStep 3 intoskills/github-agile/scripts/bump-and-tag.py, following the #163 pattern. The script runs the project's bump command, re-reads the version via--version-read-cmd, verifies thatvX.Y.Zwas created (creating an annotated fallback whentag.forceSignAnnotatedsilently rejected a lightweight tag), and pushes both the commit and the tag atomically. The resolved version is printed on stdout so the calling skill can capture it for the release step.deploy.mdStep 3 collapses from ~20 lines of narrated shell into a singlepython3 …/bump-and-tag.py …invocation per bump variant. The skill remains agnostic to whether the project uses Make, npm, uv, or a custom script — the bump and version-read commands are passed as opaque shell strings via the existing{{BUMP_*_CMD}}/{{VERSION_READ_CMD}}template variables.Issue #152