Ensure accessibility in UI components including semantic HTML, ARIA attributes, keyboard navigation, and WCAG 2.2 AA compliance.
中文简介正在准备中。当前显示原文说明。
Complete development with structured merge/PR options. Use when ready to merge or submit work.
中文简介正在准备中。当前显示原文说明。
查看安装方法安装前可以查看交给 AI 代理的指令内容。
Guide completion of development work by presenting clear options and handling chosen workflow.
Core principle: Verify tests -> Present options -> Execute choice -> Clean up.
Announce at start: "I'm using the finishing-a-development-branch skill to complete this work."
Before presenting options, verify tests pass:
# Run project's test suite
npm test / cargo test / pytest / go test ./...
If tests fail:
Tests failing (<N> failures). Must fix before completing:
[Show failures]
Cannot proceed with merge/PR until tests pass.
Stop. Don't proceed to Step 2.
If tests pass: Continue to Step 2.
# Try common base branches
git merge-base HEAD main 2>/dev/null || git merge-base HEAD master 2>/dev/null
Or ask: "This branch split from main - is that correct?"
Present exactly these 4 options:
Implementation complete. What would you like to do?
1. Merge back to <base-branch> locally
2. Push and create a Pull Request
3. Keep the branch as-is (I'll handle it later)
4. Discard this work
Which option?
Don't add explanation - keep options concise.
# Switch to base branch
git checkout <base-branch>
# Pull latest
git pull
# Merge feature branch
git merge <feature-branch>
# Verify tests on merged result
<test command>
# If tests pass
git branch -d <feature-branch>
Then: Cleanup worktree (Step 5)
# Push branch
git push -u origin <feature-branch>
# Create PR
gh pr create --title "<title>" --body "$(cat <<'EOF'
## Summary
<2-3 bullets of what changed>
## Test Plan
- [ ] <verification steps>
EOF
)"
Then: Cleanup worktree (Step 5)
Report: "Keeping branch <name>. Worktree preserved at <path>."
Don't cleanup worktree.
Confirm first:
This will permanently delete:
- Branch <name>
- All commits: <commit-list>
- Worktree at <path>
Type 'discard' to confirm.
Wait for exact confirmation.
If confirmed:
git checkout <base-branch>
git branch -D <feature-branch>
Then: Cleanup worktree (Step 5)
Run this before ANY commit, regardless of which option was chosen.
4.5.1 — Project root scan:
ls -1 | grep -vE '^(\.|node_modules|src|tests|scripts|dist|build|docs|package\.json|package-lock\.json|pnpm-lock\.yaml|tsconfig|eslint|prettier|jest|vitest|README|LICENSE|CHANGELOG|CLAUDE\.md|\.env)'
If this outputs anything, those files are AI slop. Delete or move them before proceeding.
4.5.2 — Prune stale worktrees:
git worktree prune
4.5.3 — Clean temp directory:
ls .claude/context/tmp/ 2>/dev/null && echo "Temp files exist — delete if from this session"
4.5.4 — Log to session gap log if slop was found:
echo '{"timestamp":"'$(date -u +%Y-%m-%dT%H:%M:%SZ)'","type":"cleanup","agent":"developer","description":"Deleted AI slop from project root: <filenames>","context":"finishing-a-development-branch cleanup scan"}' >> .claude/context/runtime/session-gap-log.jsonl
4.5.5 — Queue reflection if slop was found (append to reflection-spawn-request.json):
{
"id": "<uuid>",
"trigger": "ai-slop-found",
"priority": "low",
"context": "Cleanup scan found unexpected files in project root. Investigate which task created them."
}
See .claude/rules/cleanup-always.md for the full slop pattern list and correct file destinations.
For Options 1, 2, 4:
Check if in worktree:
git worktree list | grep $(git branch --show-current)
If yes:
git worktree remove <worktree-path>
For Option 3: Keep worktree.
| Option | Merge | Push | Keep Worktree | Cleanup Branch |
|---|---|---|---|---|
| 1. Merge locally | Yes | - | - | Yes |
| 2. Create PR | - | Yes | Yes | - |
| 3. Keep as-is | - | - | Yes | - |
| 4. Discard | - | - | - | Yes (force) |
Skipping test verification
Open-ended questions
Automatic worktree cleanup
No confirmation for discard
Never:
Always:
Called by:
Pairs with:
Before ANY commit on a feature/skill/agent branch:
[Unreleased] with today's datedescription: accurately reflects new capabilitiesIf any item is unchecked → DO NOT COMMIT. Update docs first.
| Anti-Pattern | Why It Fails | Correct Approach |
|---|---|---|
| Skipping test verification before merge | Broken code lands on main; CI fails after the fact | Always run test suite first; gate options on passing tests |
| Presenting open-ended completion questions | Developer doesn't know available paths; worktrees left orphaned | Present exactly 4 numbered options with clear labels |
| Deleting branch without confirmation | Developer loses in-progress work permanently | Require typed "discard" confirmation for Option 4 |
| Cleaning up worktree for Option 2 (PR) | Kills local context before PR review is complete | Only remove worktree for Options 1 and 4 |
| Merging directly without pulling latest base | Merge conflicts or stale base; CI detects drift | git pull on base branch before git merge |
Before starting:
Read .claude/context/memory/learnings.md
After completing:
.claude/context/memory/learnings.md.claude/context/memory/issues.md.claude/context/memory/decisions.mdASSUME INTERRUPTION: If it's not in memory, it didn't happen.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Ensure accessibility in UI components including semantic HTML, ARIA attributes, keyboard navigation, and WCAG 2.2 AA compliance.
中文简介正在准备中。当前显示原文说明。
Use when you want to improve response quality through meta-cognitive reasoning. Applies 15+ reasoning methods to reconsider and refine initial outputs.
中文简介正在准备中。当前显示原文说明。
N-round opposing-stance debates for trade-off analysis. Assigns pro/con roles to agents, runs structured debate rounds with quality scoring, and produces a moderator synthesis with confidence-rated recommendation. Generalizable to architecture, technology, security, and design decisions.
中文简介正在准备中。当前显示原文说明。
Force adversarial code review stance that eliminates confirmation bias — reviewer must find issues or re-analyze
中文简介正在准备中。当前显示原文说明。
Creates specialized AI agents on-demand when no existing agent matches a request. Use when the Router cannot find a suitable agent for a task. Enables self-evolution by generating persistent agents.
中文简介正在准备中。当前显示原文说明。
LLM-as-judge evaluation framework with 5-dimension rubric (accuracy, groundedness, coherence, completeness, helpfulness) for scoring AI-generated content quality with weighted composite scores and evidence citations
中文简介正在准备中。当前显示原文说明。