agentmods.dev

claude-skills: Instructions file for Claude Code

CLAUDE.md

claude-skills CLAUDE.md is an instructions file for Claude Code from yachi/claude-skills. It costs 593 tokens per session, scanned A, original, MIT.

Repository-specific instructions for working on a collection of Claude Code skills, including its layout, release steps, evaluation process, and Git workflow. A repository is a project's tracked source files and history.

In plain words
What is it for?
Use it as a project reference when changing, evaluating, or releasing a skill, and when checking that local installed files match the repository.
Why use it?
It gives a new coding agent facts that are specific to this project and cannot safely be inferred from the code alone. It helps prevent shipping an edited skill without updating the versions and installed cache it uses.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions .claude-plugin; mentions CLAUDE.md.

This is yachi/claude-skills's own configuration. It tells Claude Code how to work on claude-skills itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-skills configures →

not rated 3repo 10d ago A scan 593 tokens original MIT
View source ↗ yachi/claude-skills
Reuse

Borrowing it

Nothing to install: this file belongs to yachi/claude-skills. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/yachi/claude-skills/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/yachi/claude-skills

Made for: Claude Code.

Wrote 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.

agentmods badge for claude-skills CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/yachi/claude-skills/claude-md/github.svg)](https://agentmods.dev/instructions/yachi/claude-skills/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/yachi/claude-skills/claude-md"><img src="https://agentmods.dev/badge/instructions/yachi/claude-skills/claude-md/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.

agentmods 80×15 button for claude-skills CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/yachi/claude-skills/claude-md"><img src="https://agentmods.dev/badge/instructions/yachi/claude-skills/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 593 This file is loaded in full into every session.
When invoked 593 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00593 $0.00593
Opus 5.5 $0.00237 $0.00237
Sonnet 5.5 $0.00119 $0.00119
Haiku 4.5 $0.00059 $0.00059

Measured 9d ago against content hash ddef2b9c0e5b, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

claude-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 9d 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.

CLAUDE.md · 27 lines

How it starts

The opening of the file, as written. The whole thing — 27 lines — stays where its author put it; the contents beside it link to each section on GitHub.

claude-skills — project notes

Reference for working in this repo. It documents repo-specific facts a fresh session cannot infer. (An A/B test on 2026-10-04 found these rules did NOT change a capable agent's one-shot decisions — generic process discipline is already covered by the global CLAUDE.md and general competence. So treat this as a fast on-ramp and a checklist, not a behavior fix.)

Layout

  • skills/deep-research/ — the main skill. SKILL.md is the prompt; eval/ is its test harness; autoresearch/ is the self-improvement loop.
  • .claude-plugin/plugin.json + .claude-plugin/marketplace.json — the plugin manifests.

Shipping a skill change

  • A SKILL.md edit reaches NO installed session until the version is bumped in BOTH .claude-plugin/plugin.json AND .claude-plugin/marketplace.json. Installed sessions load the pinned cached version under ~/.claude/plugins/cache/, not the repo.
  • After the user updates the plugin, confirm the cache copy matches the repo (diff them).

Evaluating a skill change

  • eval/grade.py grades a research REPORT, not SKILL.md. Running it on the instruction file is a category error (it scores ~22-24 because the skill is not a report).
  • The only regression test is the skill executed on a prompt, then graded. Scores are high-variance: use N≥3 runs per version, and do not trust a per-eval delta or a total delta under ~3 from a single run.
  • New behavioral rules in a skill need their own eval. Sequence: build harness → run the skill live → fix graders against REAL output → THEN commit. A grader proven only on hand-authored fixtures is unverified against real inputs.
  • Authoritative graders: eval/grade_v11.py (TEXT behaviors B2,B3,B4,B6,B7,B9), eval/grade_trace.py (TRACE behaviors B1,B5,B8). eval/scenarios/grade_scenarios.py is a thin dispatcher over those two — do not re-implement grading logic elsewhere.
  • Self-test: python3 eval/selftest_v11.py (graders discriminate) and python3 eval/golden/scorecard.py (all 9 behaviors pass on real captured output).
  • When a grader disagrees with a report, read the artifact yourself and fix the grader or the skill — never the golden/report. eval/golden/ is an external oracle; never regenerate it from the graders.

Read the full file on GitHub · 27 lines

Changes

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.

  1. 9d ago First seen · 27 lines · 593 tokens per session scan A ddef2b9c0e5b

Subscribe to this mod's changes

claude-skills CLAUDE.md is an instructions file published in the GitHub repository yachi/claude-skills (3 stars, last pushed 10d ago), licensed MIT. It adds 593 tokens to every session, about $0.0024 per session on Opus 5.5. 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 2026-10-12.

Git and workflowWriting and docsai-skillsai-toolsanthropicclaudeclaude-code-plugin
Related

Other instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 9,649 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,384 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).

microsoft/vscode · 5,262 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,680 tokens

vscode copilot-instructions.md

Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).

microsoft/vscode · 2,731 tokens

deno copilot-instructions.md

Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.

denoland/deno · 3,077 tokens