Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/sangrokjung/claude-forgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/sangrokjung/claude-forge/worktree-cleanup)<a href="https://agentmods.dev/commands/sangrokjung/claude-forge/worktree-cleanup"><img src="https://agentmods.dev/badge/commands/sangrokjung/claude-forge/worktree-cleanup.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00012 | $0.01132 |
| Opus 5 | $0.00006 | $0.00566 |
| Sonnet 5 | $0.00002 | $0.00226 |
| Haiku 4.5 | $0.00001 | $0.00113 |
Grade A, and why
worktree-cleanup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task
0단계: 환경 확인
git rev-parse --git-dir
git worktree list
git branch --show-current
pwd
1단계: 정리 대상 결정
$ARGUMENTS 있으면:
- 해당 브랜치의 워크트리 찾기
$ARGUMENTS 없으면:
- 현재 디렉토리가 워크트리인지 확인
- 메인 레포면:
→ 중단❓ 정리할 워크트리를 지정하세요. 현재 워크트리 목록: [git worktree list 결과] 사용법: /worktree-cleanup feature/auth
2단계: PR 상태 확인
gh pr view --json state,mergedAt 2>/dev/null
PR이 머지됨: → 정리 진행
PR이 아직 열려있음:
⚠️ PR이 아직 머지되지 않았습니다.
PR 상태: [Open/Draft]
PR URL: [URL]
옵션:
1. "계속" - 강제 정리 (변경사항 유실 주의)
2. "취소" - PR 머지 후 다시 시도
PR 없음 (gh CLI 없거나 PR 미생성):
⚠️ PR 상태를 확인할 수 없습니다.
변경사항이 push 되었는지 확인하세요:
git log origin/[브랜치]..HEAD
옵션:
1. "계속" - 강제 정리
2. "취소" - 확인 후 다시 시도
3단계: 워크트리 정리
현재 워크트리에서 실행 중이면:
⚠️ 현재 워크트리에서는 정리할 수 없습니다.
메인 레포로 이동 후 실행하세요:
cd [메인 레포 경로]
/worktree-cleanup [브랜치명]
→ 중단
메인 레포에서 실행:
# 워크트리 제거
git worktree remove [워크트리 경로]
# 머지된 브랜치면 삭제
git branch -d [브랜치명]
# 원격 브랜치도 삭제 (머지된 경우)
git push origin --delete [브랜치명] 2>/dev/null || true
3.5단계: 체크리스트 상태 정리
워크트리 내 .claude/web-checklist-state.json 확인:
파일 있으면:
- 완료율 읽기
- 미완료 항목 있으면 경고
📋 체크리스트 상태: [completion]% ([done]/[total])
미완료 항목이 있으면:
⚠️ 미완료 체크리스트 항목 [N]개:
- [ID:N] [text]
- [ID:N] [text]
워크트리 정리 후에는 복구할 수 없습니다.
4단계: 워크트리 통계 수집
워크트리 세션 동안의 작업 통계:
# 워크트리 브랜치의 커밋 수
git log --oneline [base]...[branch] | wc -l
# 변경 파일 수
git diff [base]...[branch] --stat | tail -1
5단계: 완료 안내
════════════════════════════════════════════════════════════════
✅ Worktree Cleanup Complete (v6)
════════════════════════════════════════════════════════════════
🗑️ 삭제: [워크트리 경로]
🌿 브랜치: [브랜치명] (deleted)
📊 워크트리 통계:
커밋 수: [N]
변경 파일: [N]
추가/삭제: +[N] / -[N]
남은 워크트리:
[git worktree list 결과]
다음:
/worktree-start [새 기능] | /next-task
════════════════════════════════════════════════════════════════
에러 처리
워크트리에 커밋되지 않은 변경사항:
❌ 커밋되지 않은 변경사항이 있습니다.
[git status 결과]
옵션:
1. 변경사항 커밋 후 다시 시도
2. 강제 삭제: git worktree remove --force [경로]
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 185 lines · 12 tokens per session scan A 5f50790dbbf0
worktree-cleanup is a command published in the GitHub repository sangrokjung/claude-forge (824 stars, last pushed 4d ago), licensed MIT. It adds 12 tokens to every session and 1,132 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
deploy
Deploy application with pre/post-deploy checks.
pull-repos
Pull all repos (parent + marketplace clones + configured project repos).
changelog
Generate a changelog entry from git commits since the last tag using conventional commit format.
auto-claude
Invoke Auto-Claude autonomous coding framework for complex feature implementation.
land-and-deploy
Merge PR, wait for CI, verify deploy, run canary — the complete landing pipeline.
release-notes
Generate release notes in multiple formats from git commits.