agentic-plugins CLAUDE.md

A project instruction file for a marketplace that contains multiple Claude Code add-ons, called plugins. It explains the repository layout and the difference between the tools used by the coding agent and the tools documented by those plugins.

In plain words
What is it for?
Use it when editing plugins, skills, agent definitions, marketplace metadata, or shared development tooling in the repository. A marketplace is a catalog where these add-ons are registered and maintained.
Why use it?
It prevents contributors from putting instructions in the wrong place or referring to similarly named tools incorrectly. It also keeps project guidance separate from the plugin files that are shipped.

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/theagenticguy/agentic-plugins/claude-md
Clone the repo
git clone --depth 1 https://github.com/theagenticguy/agentic-plugins
Per session 1,213 This file is loaded in full into every session.
When invoked 1,213 The same file — it is already loaded in full.
Security scan A 0 findings. 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 $0.01213 $0.01213
Opus 5 $0.00607 $0.00607
Sonnet 5 $0.00243 $0.00243
Haiku 4.5 $0.00121 $0.00121

Measured 2d ago against content hash 4c52752683db, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentic-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 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.

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 · 106 lines

How it starts

The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.

agentic-plugins

Development context for the agentic-plugins repo. This CLAUDE.md is project-level memory for when you're editing the marketplace — it is NOT shipped as a plugin component. Plugin instructions go into SKILL.md files and agent system prompts.

Repo Structure

This repo is a Claude Code plugin marketplace at the root, hosting multiple plugins under plugins/ as sibling directories. Each plugin versions independently.

.claude-plugin/
  marketplace.json       # Marketplace registry — lists every plugin
plugins/
  agentic-skills/        # Flagship plugin
    .claude-plugin/plugin.json
    .mcp.json            # Bundled MCP servers (context7, deepwiki, brave, tavily, exa, you, awsknowledge)
    skills/              # Skill reference files (the core content)
    agents/              # Agent definitions (if present)
  plugin-gardener/       # Catalog-hygiene plugin
    .claude-plugin/plugin.json
    skills/  references/  scripts/
tools/                   # Repo-wide dev tooling (validators, scaffolding)

Harness tool names — do not confuse

Two separate tool families with similar names. Skills authored here must reference the correct one.

  • Agent — spawns a subagent. Parameters: subagent_type, prompt, description, model, run_in_background, isolation. No Stop/Output siblings — backgrounded agents auto-notify on completion. Stuck-recovery pattern is "launch a fresh Agent with a skip-completed-sections prompt," not kill-and-resume.
  • TaskCreate / TaskUpdate / TaskList / TaskGet / TaskStop — the todo-list tools. Track multi-step work, mark items in_progress / completed, wire addBlockedBy dependencies. Do NOT spawn subagents.

No roadmap or migration commentary in skill files

SKILL.md, agent definitions, and everything under references/ / templates/ ship as prompt content. Do NOT include forward-looking notes, migration history, authoring roadmap (TODO/FIXME), or meta-commentary about the skill's own state. That belongs in the PR description or a CHANGELOG — never in a file Claude reads as instructions. It costs tokens on every load and rots the moment the referenced work lands or gets dropped.

Read the full file on GitHub · 106 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 · 106 lines · 1,213 tokens per session scan A 4c52752683db

Subscribe to this mod's changes

agentic-plugins CLAUDE.md is an instructions file published in the GitHub repository theagenticguy/agentic-plugins (5 stars, last pushed 21d ago), licensed MIT. It adds 1,213 tokens to every session, about $0.0061 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.