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.
npx skills add konamgil/mandu --skill mandu-mcp-safe-changegit clone --depth 1 https://github.com/konamgil/manduWrote 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/skills/konamgil/mandu/mandu-mcp-safe-change)<a href="https://agentmods.dev/skills/konamgil/mandu/mandu-mcp-safe-change"><img src="https://agentmods.dev/badge/skills/konamgil/mandu/mandu-mcp-safe-change/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/konamgil/mandu/mandu-mcp-safe-change"><img src="https://agentmods.dev/badge/skills/konamgil/mandu/mandu-mcp-safe-change.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00070 | $0.01831 |
| Opus 5 | $0.00035 | $0.00915 |
| Sonnet 5 | $0.00014 | $0.00366 |
| Haiku 4.5 | $0.00007 | $0.00183 |
Grade A, and why
mandu-mcp-safe-change 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 11d 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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mandu MCP Safe Change
돌이키기 힘든 변경 을 실행할 때의 표준 래퍼. history.snapshot 으로 롤백 포인트를
찍고, tx.begin 으로 트랜잭션을 열고, 안에서 편집 + verify, 결과에 따라
tx.commit 또는 tx.rollback.
Trigger
- 마이그레이션: 경로 컨벤션 변경, 생성 barrel 재작성, contract 추출 등
- 대규모 리팩터: 레이어 재배치, 모듈 이동, 대량 import 경로 수정
mandu.refactor.*/mandu.refactor.migrateRouteConventions/mandu.refactor.rewriteGeneratedBarrel/mandu.refactor.extractContract호출 직전- 사용자: "돌이킬 수 없는 변경", "한 번에 바꿔줘", "전체 프로젝트에 적용"
- 생성 파일 대량 재작성 (
mandu_generate를 전체 재실행)
Recipe (순서 고정)
mandu.history.snapshot ← 롤백 포인트 (Tier-1)
└─> mandu_begin ← 트랜잭션 시작 (alias of mandu.tx.begin)
└─> <편집 / refactor tool 호출>
└─> mandu-mcp-verify fast path
├─ all green → mandu_commit (alias of mandu.tx.commit)
│ └─> mandu_prune_history (선택, 오래된 스냅샷 정리)
└─ 실패 → mandu_rollback (alias of mandu.tx.rollback)
└─> 스냅샷은 유지. 원인 분석 후 재시도
Step Detail
Step 1 — mandu.history.snapshot
mandu.history.snapshot({
repoRoot: ".",
label: "pre-route-migration-2026-04-20"
})
스냅샷은 .mandu/generated/** 같이 git 추적 밖 산출물까지 포함해서 저장한다.
label 은 항상 붙인다 — mandu_list_history 로 볼 때 뭐였는지 알아야 함.
Step 2 — mandu_begin
mandu_begin({ label: "route-convention-migration" })
트랜잭션 ID 를 기억한다. Step 4 에서 쓴다. mandu.tx.status 로 상태 조회 가능.
Step 3 — 실제 변경
이제야 파괴적 도구를 호출한다. dry-run 이 있으면 먼저 dry-run:
mandu.refactor.migrateRouteConventions({ dryRun: true, ... })
→ 변경 목록 검토
mandu.refactor.migrateRouteConventions({ dryRun: false, ... })
복합 변경이면 여러 도구를 순차 호출 — tx 안이므로 중간 실패도 안전.
Step 4 — mandu-mcp-verify fast path
변경 후 검증. mandu-mcp-verify 의 병렬 4 콜 (ate.auto_pipeline, guard_check, lint, doctor). 특히 대규모 refactor 는 lint 에러를 쉽게 새로 만들기 때문에 lint 축을 반드시 포함한다 — 하나라도 빨간 상태로 tx_commit 하지 않는다.
Step 5a — 모두 green 이면 Commit
mandu_commit({ txId })
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.
- 11d ago First seen · 165 lines · 70 tokens per session scan A 44354cc12055
mandu-mcp-safe-change is a skill published in the GitHub repository konamgil/mandu (46 stars, last pushed 15d ago), licensed MPL-2.0. It adds 70 tokens to every session and 1,831 once invoked, about $0.0003 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 skills, from other repositories
truecourse-hooks
Install, configure, or remove the TrueCourse pre-commit hook.
init-workspace-documentation
Skill "init-workspace-documentation" from griddynamics/rosetta, covering agent memory.md, agent memory, preventive rules, what worked and what failed.
truecourse-fix
Fix TrueCourse violations that have suggested fixes.
truecourse-list
List TrueCourse violations found in this repository.
plan
Research the codebase and create an implementation plan following Propose/Justify/Recommend. Use this before any feature, fix, or refactor.
validate
Run the full quality gate — tests, lint, build, and cortex-tms validate --strict.