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 agentmods add instructions/mutagent-io/skills/claude-mdgit clone --depth 1 https://github.com/mutagent-io/skillsWrote 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/instructions/mutagent-io/skills/claude-md)<a href="https://agentmods.dev/instructions/mutagent-io/skills/claude-md"><img src="https://agentmods.dev/badge/instructions/mutagent-io/skills/claude-md.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 | $0.03080 | $0.03080 |
| Opus 5 | $0.01540 | $0.01540 |
| Sonnet 5 | $0.00616 | $0.00616 |
| Haiku 4.5 | $0.00308 | $0.00308 |
Grade A, and why
skills CLAUDE.md 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 4d 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 — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for AI coding agents (Claude Code, Cursor, etc.) working inside this
repository. If you are looking for how to use the skills, read
README.md instead.
What this repo is
The public skills marketplace for the MutagenT platform. It hosts
user-facing skill bundles that ship to AI coding agents — installable via
Claude Code's /plugin marketplace add, via mutagent skills install, by
direct clone, or as a git submodule.
It satisfies two compatible specs from one source of truth:
- Agent Skills spec — every skill at
skills/<name>/SKILL.mdis a valid bare skill. - Claude Code marketplace spec
—
.claude-plugin/marketplace.jsonat the repo root wraps the skills as installable plugins.
It is not the source of truth for skill content. Skills are authored
privately (@mutagent/cli carries the canonical mutagent-cli skill in its
own source tree) and published here. Treat every file in this repo as
world-readable.
What lives here
.
├── .claude-plugin/
│ ├── marketplace.json — marketplace catalog (Claude Code reads this)
│ └── plugin.json — plugin manifest (the repo itself IS the plugin)
├── SKILL.md — source of truth (root, for crawlers/registries)
├── CHANGELOG.md
├── references/ — spec-required parent folder
│ ├── concepts/*.md — WHY/WHAT design references
│ └── workflows/*.md — HOW step sequences
├── skills/
│ └── mutagent-cli/ — byte-identical mirror (Claude Code plugin spec)
│ ├── SKILL.md — regenerated by scripts/mirror-skill.sh
│ ├── CHANGELOG.md — CI enforces parity via validate.yml mirror-check
│ └── references/ — same nested concepts/+workflows/ shape
├── CLAUDE.md — this file
├── README.md — public registry index, install paths
├── LICENSE — MIT (skills only; the CLI has its own license)
├── .github/workflows/ — validate, tag-on-merge, release
└── scripts/
├── sanitize.py — canonical public/internal filter (idempotent)
├── mirror-skill.sh — copies root content → skills/<name>/ mirror
└── sync-from-cli.sh — fetch + reconstitute skill from npm
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.
- 4d ago First seen · 276 lines · 3,080 tokens per session scan A dc9f7e21a0d2
skills CLAUDE.md is an instructions file published in the GitHub repository mutagent-io/skills (4 stars, last pushed 29d ago), licensed MIT. It adds 3,080 tokens to every session, about $0.0154 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 instructions, from other repositories
trigger_tree CLAUDE.md
Claude Code instructions for Hedde/trigger_tree: Keep this file as a router. Read only the references needed for the task.
ground CLAUDE.md
Instructions for teranos/ground, covering authority and error axiom.
CTX CLAUDE.md
Instructions for pluto2060/CTX, covering 새 ctx 정의 — 실제 production 구성 (settings.json 기준), 현재 wired hooks (production), 은퇴한 / 아카이브 hooks (wired 아님, 실행 안 됨), 공유 인프라 and g1/g2 정의 (2026-04-03 명료화).
radin AGENTS.md
Instructions for shortcuts/radin, covering radin — agent reference, project, dev loop, storage contract and backlog entry schema.
radin CLAUDE.md
Instructions for shortcuts/radin, a project described as: 🐀 the agentic stack for those who have to save tokens.
river-review GEMINI.md
Gemini CLI instructions for s977043/river-review, covering gemini instructions (river-review), 必須チェック, gemini-specific tips, prompt assembly and development tips.