From vibe-center-skills
Performs PR closeout: confirms merge, closes owned issues, creates follow-up tickets, deletes local/remote branches and worktrees, switches to main, and appends handoff record.
How this skill is triggered — by the user, by Claude, or both
Slash command
/vibe-center-skills:vibe-doneThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
PR 合并后的完整收口流程。
PR 合并后的完整收口流程。
Use vibe-done for human-initiated manual cleanup after PR merge.
Use vibe-closeout only when Manager has written handoff indicate with cleanup instructions.
gh pr view <pr-number> --json state,mergedAt
/vibe-integrate检查当前 flow 负责的 issue 是否已关闭:
gh issue view <issue-number> --json state
判断是否需要创建 follow-up issue:
检查点:
若需要 follow-up:
gh issue create --title "Follow-up: <主题>" --body "来源: #<原 issue>, PR #<pr-number>"
只关闭当前 flow 负责的主 issue:
gh issue close <issue-number> --comment "已完成 in PR #<pr-number>"
不关闭关联 issue(由各自 flow 负责)。
删除本地和远程 branch:
# 删除本地 branch
git branch -D <branch-name>
# 删除远程 branch
git push origin --delete <branch-name>
删除当前 worktree(资源清理):
# 先确认不在要删除的 worktree 中
pwd
# 删除 worktree
git worktree remove <worktree-path>
git checkout main
git pull origin main
在当前 flow 的 handoff 链中记录终态:
vibe3 handoff append "vibe-done: flow closed" --actor vibe-done --kind note
格式:
## Flow Closure
- flow: <flow-name>
- status: completed
- pr: <pr-link> (merged)
- issues_closed: <issue-links>
- follow_up: <issue-link or none>
- resources_cleaned: branch + worktree
- completed_at: <ISO-8601>
完成后输出:
npx claudepluginhub jacobcy/vibe-coding-control-centerProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.