Borrowing it
Nothing to install: this file belongs to WeZZard/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.
curl -O https://raw.githubusercontent.com/WeZZard/skills/main/CLAUDE.mdgit clone --depth 1 https://github.com/WeZZard/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/wezzard/skills/claude-md)<a href="https://agentmods.dev/instructions/wezzard/skills/claude-md"><img src="https://agentmods.dev/badge/instructions/wezzard/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.
<a href="https://agentmods.dev/instructions/wezzard/skills/claude-md"><img src="https://agentmods.dev/badge/instructions/wezzard/skills/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00965 | $0.00965 |
| Opus 5.5 | $0.00386 | $0.00386 |
| Sonnet 5.5 | $0.00193 | $0.00193 |
| Haiku 4.5 | $0.00097 | $0.00097 |
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 2mo 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 — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Add a new plugin to .claude-plugin/marketplace.json and merge it. The Auto Register Plugin Website workflow runs the deterministic registration script for plugins absent from catalog/website-registry.json, then opens a registration pull request. The registration script does not use an agent. Merging that pull request creates a site-building:queued issue. The queued worker uses Pi to generate missing or stale website copy, then opens a separate website-content pull request. To opt a plugin out, add its registry entry with website: false before merging. Manual dispatch (gh workflow run register-plugin-website.yml -f plugin=<name>) remains available for backfill.
The plugin root defines publication. Every skills/<name>/SKILL.md file is an installable skill and appears on the website. A skill under .agents/skills/<name>/SKILL.md is repository-only: plugin installation and website generation exclude it. Move maintenance and development skills into .agents/skills/ instead of asking Pi to classify or hide them.
Website TOML lives in catalog/website/<plugin>.{plugin,skills,philosophy}.toml. The catalog copy wins. A website.*.toml file inside the plugin repository is a legacy fallback that scripts/update-plugin-website.mjs migrates during site building. Pi generates missing entries only for published plugin-root skills and persists them in the site-building pull request. The script removes TOML entries and plugin-owned generated JSON when a published skill disappears.
Machine-generated entries contain source_hash, which fingerprints their SKILL.md, and content_hash, which fingerprints their generated copy. An unchanged source produces no website change. A changed source regenerates a machine-owned entry through Pi. The worker preserves an entry whose current copy no longer matches content_hash because that mismatch identifies a human edit. Entries without hashes also count as human-edited. Delete an entry to let Pi regenerate it, or run node scripts/adopt-plugin-content.mjs --plugin <name> to mark its current copy as machine-owned without changing the copy.
The site-building worker uses Pi 0.81.1, DeepSeek deepseek-v4-pro, and thinking level high. The worker loads only .pi/skills/site-building/SKILL.md and disables tools, sessions, global skills, extensions, prompt templates, and context files. GitHub Issues store durable tasks with site-building:queued, site-building:running, site-building:failed, and site-building:blocked labels. Every enqueue creates an immutable issue; the serialized worker deduplicates requests for the same plugin and recovers a stale running task after an interrupted run. Only issues carrying one of those status labels participate in the queue. The worker runs only outside DeepSeek peak pricing and the hour before each peak period. Preview and deployment workflows render committed content without invoking an agent.
Catalog merges complete plugin releases before site building starts. catalog-sync-notify.yml sends the existing release callback first, then enqueues website content when the plugin has website registration. Site-building failures therefore cannot hold a plugin release open.
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.
- 2mo ago First seen · 40 lines · 965 tokens per session scan A 85eccd8aca67
skills CLAUDE.md is an instructions file published in the GitHub repository WeZZard/skills (25 stars, last pushed 1mo ago), licensed MIT. It adds 965 tokens to every session, about $0.0039 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 2026-08-30.
skills CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| skills CLAUDE.mdWeZZard/skills | Instructions file | 25 | 1mo ago | MIT | A | 965 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 5,516 | GitHub Copilot |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | today | MIT | A | 4,680 | Codex, OpenCode |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | today | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 275k | 3d ago | MIT | A | 4,313 | Codex, OpenCode |