pm-guard CLAUDE.md

pm-guard CLAUDE.md is an instructions file for coding agents from udonc/pm-guard. It costs 644 tokens per session, scanned A, original, MIT.

Projektregeln für `pm-guard`, ein Claude-Code-Plugin, das falsche Paketmanager-Befehle blockiert. Ein Paketmanager wie npm, yarn oder pnpm installiert Abhängigkeiten und führt Projektbefehle aus.

In plain words
What is it for?
Zum Festlegen des erlaubten Paketmanagers und zum Prüfen von Bash-Befehlen vor ihrer Ausführung. Die Anleitung beschreibt auch Plugin-Dateien, Hooks und Tests.
Why use it?
Es verhindert, dass ein Coding-Agent im Projekt den nicht erlaubten Paketmanager verwendet. Das reduziert Fehler durch unterschiedliche Abhängigkeiten oder Skripte.

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/udonc/pm-guard/claude-md
Clone the repo
git clone --depth 1 https://github.com/udonc/pm-guard

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 pm-guard CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/udonc/pm-guard/claude-md.svg)](https://agentmods.dev/instructions/udonc/pm-guard/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/udonc/pm-guard/claude-md"><img src="https://agentmods.dev/badge/instructions/udonc/pm-guard/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 644 This file is loaded in full into every session.
When invoked 644 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 $0.00644 $0.00644
Opus 5 $0.00322 $0.00322
Sonnet 5 $0.00129 $0.00129
Haiku 4.5 $0.00064 $0.00064

Measured 4d ago against content hash a301dd112cb8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pm-guard 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 4d 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 · 44 lines

How it starts

The opening of the file, as written. The whole thing — 44 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.

Project Overview

pm-guard is a Claude Code plugin that prevents Claude from using the wrong package manager (npm/yarn/pnpm/bun/deno). It works as a PreToolUse hook that intercepts Bash tool calls and blocks commands using disallowed package managers.

Architecture

This is a Claude Code plugin distributed as a custom marketplace. The repo root is the marketplace; the plugin lives under plugins/pm-guard/:

  • .claude-plugin/marketplace.json - Marketplace manifest (name, owner, plugin list)
  • plugins/pm-guard/.claude-plugin/plugin.json - Plugin manifest (name, version, metadata)
  • plugins/pm-guard/hooks/hooks.json - Hook registration: binds check-pm.sh to PreToolUse events on the Bash tool
  • plugins/pm-guard/hooks/check-pm.sh - Core logic: a pure bash script that reads tool input JSON from stdin, extracts the command, detects the allowed package manager, and blocks disallowed ones

How check-pm.sh Works

  1. Parses stdin JSON to extract tool_input.command (using sed + awk, no jq dependency)
  2. Detects allowed PM with priority: PM_GUARD_ALLOWED env var > package.json packageManager field > lockfile detection
  3. Strips quoted content (single and double quotes) to avoid false positives on strings like grep "npm" package.json. Known trade-off: bash -c "npm install" won't be caught.
  4. Blocks disallowed PMs using word-boundary regex (not preceded by [a-zA-Z0-9_.-], not followed by [a-zA-Z0-9_./-]) to avoid false positives like pnpm-lock.yaml, .npm/, npm-check
  5. Outputs hook JSON with one of three outcomes:
    • deny decision with reason (blocked PM detected)
    • systemMessage warning (PM could not be detected AND command contains a PM keyword)
    • Clean exit with no output (command allowed, or no PM detected and command has no PM keywords)

Development

# Test the plugin locally
claude --plugin-dir ./plugins/pm-guard

# Run check-pm.sh directly with mock input
echo '{"tool_input":{"command":"npm install foo"}}' | PM_GUARD_ALLOWED=pnpm ./plugins/pm-guard/hooks/check-pm.sh

# Run BATS tests
./scripts/setup-tests.sh
./tests/test_helper/bats-core/bin/bats tests/check-pm.bats

Read the full file on GitHub · 44 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. 4d ago First seen · 44 lines · 644 tokens per session scan A a301dd112cb8

Subscribe to this mod's changes

pm-guard CLAUDE.md is an instructions file published in the GitHub repository udonc/pm-guard (3 stars, last pushed 5mo ago), licensed MIT. It adds 644 tokens to every session, about $0.0032 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.

Related

Other instructions, from other repositories

claude-skills CLAUDE.md

Claude Code instructions for secondsky/claude-skills, covering claude skills - project context, ⚠️ critical: communication rule — the most important rule, ⚠️ critical: repository policy - never touch jezweb repo, ❌ forbidden actions (jezweb repo) and ✅ required actions (secondsky repo).

secondsky/claude-skills · 7,874 tokens

claude-skills AGENTS.md

AGENTS.md instructions for secondsky/claude-skills, covering agent instructions, ⚠️ critical: communication rule — the most important rule and full project context.

secondsky/claude-skills · 154 tokens

ccx-context-system CLAUDE.md

Claude Code instructions for shck-dev/ccx-context-system, covering claude.md, what this is, commands, architecture and load-bearing seams.

shck-dev/ccx-context-system · 1,059 tokens

dot-claude-plugins CLAUDE.md

Instructions for and3r817/dot-claude-plugins, covering claude.md, repository overview, core architecture principles, testing commands and run all test suites (master runner).

and3r817/dot-claude-plugins · 1,578 tokens

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 · 6,785 tokens

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.

github/spec-kit · 7,104 tokens