From speckit-claude
Monitor CI/CD deployment, check Copilot review notes, evaluate findings, and offer to fix. Run after merging a PR.
How this skill is triggered — by the user, by Claude, or both
Slash command
/speckit-claude:deploy-monitorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
```text
$ARGUMENTS
Run this skill after merging a PR. It monitors the deployment and handles Copilot feedback automatically.
If $ARGUMENTS contains a PR number, use that. Otherwise, find the most recently merged PR:
gh api repos/{owner}/{repo}/pulls?state=closed&sort=updated&direction=desc --jq '.[0] | .number'
Find the most recent active workflow run for the PR branch:
gh run list --branch "$(gh pr view {pr} --json headRefName --jq '.headRefName')" \
--limit 5 --json databaseId,status,conclusion,name \
--jq '.[] | select(.status != "completed") | .databaseId' | head -1
If an active run is found, subscribe to it — this blocks and streams live output until the run completes:
gh run watch <run-id> --exit-status
If no active run is found (already completed), fetch the most recent result on the same branch:
gh run list --branch "$(gh pr view {pr} --json headRefName --jq '.headRefName')" --limit 1 \
--json status,conclusion,name --jq '.[0] | .name + ": " + .conclusion'
--exit-status exits non-zero if the run fails — no manual status check neededgh run view <run-id> --web) and stop. Suggest investigation.Fetch Copilot review comments on the PR:
gh api repos/{owner}/{repo}/pulls/{pr}/comments --jq '.[] | .path + ":" + (.line // .original_line | tostring) + " | " + .body'
Also check the review summary:
gh api repos/{owner}/{repo}/pulls/{pr}/reviews --jq '.[] | select(.user.login | contains("copilot")) | .state + " | " + .body'
If there are 0 Copilot comments:
If there are Copilot comments:
After presenting findings, ask the user:
"Copilot found {N} issues. Want me to fix all of them?"
Options:
If fixing:
fix: {N} Copilot findings — {summary}Final report:
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub gratefuljinx77/speckit-claude --plugin speckit-claude