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/smk-labs/claude-plugins/claude-mdgit clone --depth 1 https://github.com/smk-labs/claude-pluginsWrote 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/smk-labs/claude-plugins/claude-md)<a href="https://agentmods.dev/instructions/smk-labs/claude-plugins/claude-md"><img src="https://agentmods.dev/badge/instructions/smk-labs/claude-plugins/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 5.1 | $0.00787 | $0.00787 |
| Opus 5 | $0.00394 | $0.00394 |
| Sonnet 5 | $0.00157 | $0.00157 |
| Haiku 4.5 | $0.00079 | $0.00079 |
Grade A, and why
claude-plugins 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 5d 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
claude-plugins
The marketplace repo for my plugins. One plugin per folder at the repo root, listed once in .claude-plugin/marketplace.json. There is no plugins/ directory, and a path that assumes one 404s silently: readable's CDN line carried @main/plugins/readable/assets/rc.css for a year and only ever showed up as an unstyled card.
One source, always the remote
Every machine and every Claude surface installs these plugins from the remote marketplace, https://github.com/smk-labs/claude-plugins.git. A second copy anywhere means two versions, and the stale one wins half the time.
So never leave any of these behind:
- a skill copied into
~/.claude/skills/<name>/or a command into~/.claude/commands/<name>.mdwhen a plugin here already ships it - a marketplace registered with
source: directorypointing at this checkout claude plugin installfrom a local path- a dev override in the Claude desktop config (for readable, an
mcpServers.readable-cardentry whose path points into this checkout instead of~/.claude/plugins/data/readable/server/server.js)
A plugin here MAY register itself with the desktop app on first session, because installing a plugin should be the whole setup. readable does. But an auto-write earns that only by satisfying all four of these, and readable's hooks/connect.sh is the worked example plus the tests:
- every profile, not one. A hook that knows a single config path leaves the other profiles diverged, which is what drives people to hand-copy whole configs between profiles and spread live API tokens doing it.
- one backup per config, ever. Not one per write, or a config full of tokens accumulates copies of itself.
- say it once. The write is useless until the app restarts, so a silent write leaves a user with no feature and no explanation.
- an opt-out that holds. Removal must survive the next session, or removal is not a thing the user can do.
Plus the precondition that makes it safe at all: being registered must not imply being able to act. readable's server refuses to offer its card tool to a host that cannot render one, so registering it in a profile that cannot use it costs nothing.
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.
- 5d ago First seen · 42 lines · 787 tokens per session scan A 32292bfd0c35
claude-plugins CLAUDE.md is an instructions file published in the GitHub repository smk-labs/claude-plugins (11 stars, last pushed 4d ago), licensed MIT. It adds 787 tokens to every session, about $0.0039 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 instructions, from other repositories
claude-plugins AGENTS.md
Instructions for jontsai/claude-plugins, covering agent guidelines, project overview, key constraints, plugin structure and validation.
PM-Copilot-by-Product-Faculty CLAUDE.md
Claude Code instructions for Productfculty-aipm/PM-Copilot-by-Product-Faculty, covering pm copilot — global instructions, session start protocol, memory update protocol, context loading (all pm operations) and framework defaults.
claude-code-plugins CLAUDE.md
Claude Code instructions for qte77/claude-code-plugins, a project described as: A Claude Code plugin marketplace providing skills, rules, and scripts extracted from a production development workflow.
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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
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.