agent-plugins CLAUDE.md

A set of repository instructions for Claude Code plugins in a public plugin marketplace. It describes plugin layout, required files, conventions, and checks for accidentally publishing personal information.

In plain words
What is it for?
Working on plugin folders, marketplace registration, documentation, installation links, required guidance files, and pre-commit privacy checks.
Why use it?
It gives contributors the local rules needed to add or change plugins safely and keep the marketplace structure consistent.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/trtmn/agent-plugins/claude-md
Clone the repo
git clone --depth 1 https://github.com/trtmn/agent-plugins
Per session 965 This file is loaded in full into every session.
When invoked 965 The same file — it is already loaded in full.
Security scan B 1 finding. Scan, not verified.
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 $0.00965 $0.00965
Opus 5 $0.00483 $0.00483
Sonnet 5 $0.00193 $0.00193
Haiku 4.5 $0.00097 $0.00097

Measured 2d ago against content hash 27a5e7021d5d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

agent-plugins CLAUDE.md scanned grade B with 1 finding 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 2d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- `.gitignore` — excludes `**/.claude/settings.local.json`, `**/evals/`, workspace dirs, `.DS_Store`
CLAUDE.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Overview

A Claude Code plugin marketplace. Plugins live under plugins/; .claude-plugin/marketplace.json at the repo root registers them all.

Install path:

/plugin marketplace add trtmn/agent-plugins
/plugin install <plugin-name>@agent-plugins

Plugin Directory Structure and Conventions

See AGENTS.md — it is the canonical source for the required plugin directory layout, plugin.json/SKILL.md/CHANGELOG.md conventions, the $CLAUDE_PLUGIN_ROOT path convention (and its exceptions), and when a plugin needs a setup command. Read it before creating or modifying any plugin.

PII Prevention — applies to every commit, not just plugin work

This repo mirrors to a public GitHub repo. Before committing anything — plugin code, docs/, root-level files — see AGENTS.md's "PII Prevention" section and run its mandatory pre-commit sweep. It has already happened once (a personal Tailscale hostname, SSH username, and machine names leaked via docs/ and a plugin, requiring a git filter-repo history rewrite across every branch). docs/ is not exempt — everything in this repo gets committed and pushed the same way.

Existing Plugins

cowsay, font-extractor, home-assistant, homebrew-dev, imsg, jellyfin, mastodon-cli, obsidian-cli, preflight-check, pushover, quack, recipe-fetch, self-improvement, side-quest, skills-manager, tailscale-policy-manager, touch_file, unifi-api, video-extract, wp-custom-theme, wp-local, youtube-data-api, yt-dlp.

self-improvement is the full learning loop in one plugin (as of v2.0.0 it absorbed the former standalone learnings plugin). It ships two skills + three agents:

  • Capture (learnings skill + agent) — autonomous. Main Claude delegates to the learnings subagent in the background on every correction/error/suggestion; it appends Status: pending entries to ~/.learnings/.
  • Review + auto-promote (self-improvement skill + agent, plus the learning-investigator agent) — autonomous and manual. A gated SessionEnd hook spawns a detached headless claude -p review; /self-improvement runs the same pipeline foreground. The learning-investigator judges each entry against a conservative bar and the orchestrator auto-promotes qualifiers (user-scope) into CLAUDE.md, logging a revertible trail to ~/.learnings/CHANGELOG.md. Undo with /self-improvement:revert <PROMO-hex>. Wire it up with /self-improvement:setup.

Read the full file on GitHub · 53 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. 2d ago First seen · 53 lines · 965 tokens per session scan B 27a5e7021d5d

Subscribe to this mod's changes

agent-plugins CLAUDE.md is an instructions file published in the GitHub repository trtmn/agent-plugins (2 stars, last pushed 5d ago), licensed Unlicense. It adds 965 tokens to every session, about $0.0048 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.