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-orientgit 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-orient)<a href="https://agentmods.dev/skills/konamgil/mandu/mandu-mcp-orient"><img src="https://agentmods.dev/badge/skills/konamgil/mandu/mandu-mcp-orient/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-orient"><img src="https://agentmods.dev/badge/skills/konamgil/mandu/mandu-mcp-orient.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 | $0.00072 | $0.01442 |
| Opus 5 | $0.00036 | $0.00721 |
| Sonnet 5 | $0.00014 | $0.00288 |
| Haiku 4.5 | $0.00007 | $0.00144 |
Grade A, and why
mandu-mcp-orient 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 10d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mandu MCP Orient
세션에 처음 진입하거나 긴 휴지 뒤 복귀할 때 "지금 프로젝트가 어떤 상태인지" 를 한 번에 파악하는 레시피. 개별 진단 도구를 나열하지 않고 집계부터 시작한다.
Trigger
- "현 상태?", "지금 어디까지 됐지?", "무슨 일 있었어?"
- "여기서부터 계속", 세션 재진입
- 긴 대화 복귀,
/sc:load직후 - 사용자가 프로젝트 이름만 말하고 추가 지시가 없을 때
Recipe (순서 고정)
mandu.ai.brief ← Tier-0 aggregate: 먼저 호출
└─> mandu_get_architecture ← 프리셋/레이어 규칙
└─> mandu_kitchen_errors ← 런타임 에러 스트림 (있으면 최근 N개)
└─> mandu_guard_check ← snapshot only, heal 금지
Step 1 — mandu.ai.brief
mandu.ai.brief({ depth: "short" })
반환 필드:
title,summary— 프로젝트 정체성skills— 설치된 @mandujs/skills + 프로젝트별 generated skillsrecent_changes— 최근 20 커밋 subject/hashdocs—docs/최상위 목차config.guard_preset,config.fs_routes,config.has_playwrightsuggested_next_steps— 최근 활동 기반 추천
read-only, fail-soft. git 이나 docs 가 없어도 에러 나지 않음.
depth 는 상황에 맞춰:
- 처음 진입:
"short"으로 충분 - 장기 복귀 / 복잡 프로젝트:
"full"
Step 2 — mandu_get_architecture
mandu_get_architecture({})
alias of mandu.brain.architecture. 프리셋 (mandu / fsd / clean / hexagonal / atomic / cqrs),
레이어 방향, 허용/금지 import 규칙을 반환. ai.brief 로 "무엇을 하는 프로젝트인지" 는
알았으니, 여기서 "어떤 규칙으로 돼 있는지" 를 확인한다.
Step 3 — mandu_kitchen_errors
mandu_kitchen_errors({ limit: 10 })
dev 서버가 실행 중이면 최근 런타임 에러를 스트림에서 가져온다. 빌드는 됐지만
실행 중 에러가 쌓여 있는 경우 — ai.brief / get_architecture 만으로는 안 보인다.
dev 서버가 없으면 빈 배열. 실패 아니다.
Step 4 — mandu_guard_check (snapshot only)
mandu_guard_check({ repoRoot: "." })
CRITICAL: heal 하지 않는다. 오리엔테이션 단계에서는 현황만 본다.
위반이 있으면 그 사실을 기록하고 사용자에게 알리되, 고치는 건
mandu-mcp-verify 나 mandu_guard_heal 을 명시적으로 호출할 때만.
종료 조건
4 단계가 끝나면 다음 중 하나로 전이:
| 결과 | 다음 skill |
|---|---|
| 사용자가 "만들어줘" / 새 피처 요청 | mandu-mcp-create-flow |
| guard 위반 존재 + 사용자가 정리 원함 | mandu-mcp-verify -> drill-down |
| 에러 있음 + 사용자가 디버그 원함 | mandu-debug |
| 빌드/배포 원함 | mandu-mcp-deploy |
사용자 지시가 없으면 suggested_next_steps 요약 제시 |
— |
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.
- 10d ago First seen · 129 lines · 72 tokens per session scan A 084864a94352
mandu-mcp-orient is a skill published in the GitHub repository konamgil/mandu (46 stars, last pushed 14d ago), licensed MPL-2.0. It adds 72 tokens to every session and 1,442 once invoked, about $0.0004 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
familiar-help
Use familiar companion tools for creative buddy/companion interactions. Handles fortunes, code roasts, haiku, stats, mood, lore, focus timers, and personality changes. Triggers when user addresses their companion by any name — "buddy" (Anthropic's UI label), "familiar", "companion", or whatever name the user uses for…
trce-share
Share the current Claude Code session with trce after an explicit user request.
rust-project
Modern Rust project architecture guide for 2025. Use when creating Rust projects (CLI, web services, libraries). Covers workspace structure, error handling, async patterns, and idiomatic Rust best practices.
slides
A slide generator that creates PNG images for presentation decks, including simple backgrounds for spoken-video scripts and fuller layouts for standalone presentations.
typescript-project
Modern TypeScript project architecture guide for 2025. Use when creating new TS projects, setting up configurations, or designing project structure. Covers tech stack selection, layered architecture, and best practices.
phone-call
Place outbound AI phone calls via Bland AI: book reservations, make appointments, ask businesses questions, then report back the transcript and summary.