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 tigu77/tiguclaw --skill claude-wrapper-syncgit clone --depth 1 https://github.com/tigu77/tiguclawWrote 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/tigu77/tiguclaw/claude-wrapper-sync)<a href="https://agentmods.dev/skills/tigu77/tiguclaw/claude-wrapper-sync"><img src="https://agentmods.dev/badge/skills/tigu77/tiguclaw/claude-wrapper-sync.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 | $0.00135 | $0.02047 |
| Opus 5 | $0.00068 | $0.01024 |
| Sonnet 5 | $0.00027 | $0.00409 |
| Haiku 4.5 | $0.00014 | $0.00205 |
Grade A, and why
claude-wrapper-sync 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 3d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
claude-wrapper-sync — Claude Code 자산을 tiguclaw 로 래핑
Claude Code 의 .claude/{agents,skills,commands} 를 tiguclaw 의 {agents,skills,commands} 로
가져온다. 세 포맷이 사실상 동일(agents=<name>.md frontmatter name/description/model/tools,
skills=<dir>/SKILL.md frontmatter name/description)이라 래핑은 검증 + 복사가 본질이다.
새 코어·라이브러리 없음. tiguclaw 는 이 자산을 매 턴 라이브 발견하므로 재시작 불요 —
동기화 직후 find_skills/find_agents 또는 바로 invoke_skill/spawn_agent 로 쓸 수 있다.
0) ★메인 지침 먼저 이해·반영 (CLAUDE.md → PROJECT.md)
Claude Code 스킬/에이전트는 프로젝트 CLAUDE.md(+ 하위 **/CLAUDE.md, 전역 ~/.claude/CLAUDE.md)가
세운 컨벤션·아키텍처·규칙 컨텍스트를 전제한다. 스킬만 복사하면 그 맥락이 없어 겉돈다. 순서:
- 먼저
<project>/CLAUDE.md(및 하위·전역)를 Read 해 이해한다. - 그 핵심 컨텍스트를 tiguclaw 의
<project>/PROJECT.md에 반영한다 — CLAUDE.md ↔ tiguclaw PROJECT.md (둘 다 프로젝트 메인 지침). PROJECT.md 가 없으면 요지를 정리해 작성(frontmattername·description·status: active), 있으면 덮어쓰지 말고 스킬이 의존하는 규칙만 보강. ★verbatim 통째 복사보다 요지(스킬이 전제하는 컨벤션·용어·제약)를 담아라(길면 요약). - ★LLM-agnostic(#2): CLAUDE.md 는 claude SDK 만 자동 로드하므로, PROJECT.md 에 반영해야 codex/openai 어댑터로도 같은 맥락이 선다. SDK 자동로드에 의존하지 말 것.
- 컨텍스트에 의존하는 스킬은 홈 전역보다 프로젝트 스코프(
<project>/.tiguclaw/skills) 래핑 권장 (맥락 PROJECT.md 와 같은 프로젝트에서 함께 산다). 그 폴더는project_register로 등록.
1) 소스·대상 정하기
- 소스: 사용자가 준 Claude Code 프로젝트 경로의
<project>/.claude/(없으면~/.claude/). 하위:.claude/agents/*.md,.claude/skills/<name>/SKILL.md,.claude/commands/*.md. - 대상 (둘 중, 불명확하면 사용자에게 물어라):
- 전역 재사용 →
<TIGUCLAW_HOME>/{agents,skills,commands}(기본. 모든 대화서 사용). - 특정 프로젝트 전용 → 그 프로젝트 폴더의
.tiguclaw/{agents,skills,commands}(cwd 스코프, 2026-07-10 메타 폴더 컨벤션 — 프로젝트 자기 폴더와 충돌 회피). 레거시 flat<project>/skills도 발견되나 신규는.tiguclaw/아래에 쓴다.
- 전역 재사용 →
Glob로 소스 목록을 먼저 만든다.
2) 드라이런 미리보기 (먼저 보여주고 확인)
쓰기 전에 무엇이 래핑될지 목록을 만들어 사용자에게 보인다:
- 각 항목: 종류(agent/skill/command)·이름·소스경로 → 대상경로.
- 이름 충돌: 대상(또는 tiguclaw 빌트인 harness·code-review 등)에 같은 이름이 이미 있으면 ⚠️ 표시하고 덮어쓰기 여부를 명시 확인받는다(빌트인 클로버링 금지 — 스킵/개명 권장).
- frontmatter
name·description누락 항목은 드롭(tiguclaw 파서가 버림) — 목록에 "드롭(사유)"로.
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.
- 3d ago Changed · +1 lines 93b012487155
- 8d ago First seen · 77 lines · 135 tokens per session scan A 57005d55c0d3
claude-wrapper-sync is a skill published in the GitHub repository tigu77/tiguclaw (6 stars, last pushed today), licensed Apache-2.0. It adds 135 tokens to every session and 2,047 once invoked, about $0.0007 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 2026-08-31.
Other skills, from other repositories
commonly
You are a member of a Commonly workspace — a shared space where humans and AI agents from any origin collaborate in pods (chat rooms with memory). Use this whenever you are connected to Commonly via the commonly MCP tools: to read what's happening, post, remember things across sessions, react, DM other agents, and…
github
Interact with GitHub (issues, PRs, repos, releases) using the gh CLI. Use when asked to read or write GitHub state — open an issue, fetch PR diff, comment, list runs, etc.
pandic-office
Convert Markdown to PDF (or DOCX/EPUB/HTML) using the pandoc CLI. Use when asked to produce a PDF report, brief, summary, or any document where the input is Markdown and the output should be a polished, paginated file.
Manipulate PDF files — extract text, count pages, render thumbnails, merge or split documents. Use for PDF-specific operations that don't fit markdown-converter (general read) or pandic-office (write from markdown).
tmux
Manage long-running shell sessions with tmux — start a detached session, run a long task, reattach later, capture output. Use when a task takes longer than a single tool call (build, test, log tail).
community-research-insight
Extract structured insight briefs from community research transcripts or notes. Produces pain points, stakeholder needs, opportunity maps, risks, and follow-up questions. Requires human review before publication.