원클릭으로
create-pr
Run /verify-pr checks, then create a GitHub PR if all pass. Automates the full flow from quality checks to PR creation.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Run /verify-pr checks, then create a GitHub PR if all pass. Automates the full flow from quality checks to PR creation.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | create-pr |
| description | Run /verify-pr checks, then create a GitHub PR if all pass. Automates the full flow from quality checks to PR creation. |
| argument-hint | [--base <branch>] |
Run all quality checks and create a GitHub PR if everything passes.
Ensure on a feature branch with the right cwd:
git branch --show-current — check current branchmain (or master), also run git worktree list --porcelain and check whether any other worktree is on a non-main branch. If yes, surface that explicitly:
git worktree list --porcelain | awk '/^worktree /{wt=$2} /^branch refs\/heads\//{b=substr($0,index($0,"refs/heads/")+11); if (b!="main" && b!="master") print wt " on branch " b}'
The user almost certainly intended to run /create-pr from that worktree's path — every subsequent git / gh pr view call defaults to cwd, so running from the parent worktree on main produces "no PR found for branch 'main'" even when a feature branch with commits is sitting in another worktree. Stop and ask the user whether to proceed in the listed worktree (offering cd <path> as the next step) before falling through to "create a new branch".main and no other worktrees have non-main branches, ask the user for a branch name and create it: git checkout -b <branch-name>.feat/, fix/, refactor/, docs/, chore/ prefix.Run /verify-pr — typecheck, lint, build, tests, CI, docs consistency, leftover resources. If any check fails, stop and report.
Ensure changes are committed and pushed:
git status — if uncommitted changes exist, ask the user whether to commit themgit push -u origin <branch> — ensure remote is up to dateCheck if PR already exists for the current branch:
gh pr view --json number,url -q '.url' 2>/dev/nullGather PR context:
git log main...HEAD --oneline — all commits in this branchgit diff main...HEAD --stat — changed files summarymain, overridable with --base)Draft PR title and body:
git diff main...HEAD), not just commit messages — commit messages may reflect intermediate iterations that were later reverted## Summary
- bullet points of what changed and why
## Test plan
- [ ] Unit tests pass (N files, M tests)
- [ ] Integration test: <which ones were run, if any>
- [ ] Documentation updated
Create PR:
gh pr create --title "..." --body "$(cat <<'EOF'
...
EOF
)"
Report the PR URL.
/verify-pr check failsgh pr edit may fail silently (e.g., Projects Classic deprecation). After updating a PR, verify the result with gh pr view. If gh pr edit fails, fall back to gh api repos/{owner}/{repo}/pulls/{number} -X PATCHDetect and delete leftover AWS resources from cdkd integration tests. Only targets resources matching known cdkd stack name patterns.
Proactively hunt for cdkd bugs by deploying real CDK apps that exercise common-but-untested AWS resources, configs, and CloudFormation notations against real AWS, then fix what breaks. Use for a periodic "find latent bugs" sweep, not for verifying a specific change.
Work through already-filed GitHub issues (typically the bug-hunt's output) end to end — triage safely, pick a few FILE-DISJOINT issues to fix in parallel, claim each on the issue before starting (collision-safe with other agents), verify against real AWS, then carry each through merge → pull → release → rebuild the linked binary → worktree cleanup. Use when asked to "handle/address filed issues", not to hunt for new bugs (that is /hunt-bugs).
Recommend which integration tests to run, based on the integ ledger (staleness / last result) plus the code areas touched by recent commits. Outputs a prioritized list of `/run-integ <name>` commands. Use before a release, after a batch of merges, or when unsure what integ coverage a change needs.
Scaffold a new integration test for cdkd. Creates a minimal CDK app with the specified AWS resources for deploy/destroy E2E testing.
Run integration tests (deploy + destroy) against real AWS. Use when you need to verify cdkd works end-to-end with actual AWS resources.