A command that checks whether a Claude Code project, its tools, and its supporting files follow the expected setup. It examines folders, settings, hooks, permissions, memory files, and installed dependencies.

Use it for a full infrastructure check or to focus on framework setup or tool installation; the fix option can apply detected corrections.

It helps find missing files, invalid configuration, broken references, permission problems, and other setup issues before they cause trouble.

audit — Command by stefan-jansen

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 commands/stefan-jansen/claude-code-toolkit/audit

Clone the repo

git clone --depth 1 https://github.com/stefan-jansen/claude-code-toolkit

Per session 6 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,193 The whole file, excluding the scripts and references it only reads on demand.

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.

Model	Per session	Once invoked
Fable 5	$0.00006	$0.01193
Opus 5	$0.00003	$0.00596
Sonnet 5	$0.00001	$0.00239
Haiku 4.5	$0.00001	$0.00119

Measured yesterday against content hash 16f51388a4f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

audit 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 yesterday.

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.

plugins/system/commands/audit.md · 175 lines

How it starts

The opening of the file, as written. The whole thing — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Framework Infrastructure Audit

Validates Claude Code framework setup, tool installation, and infrastructure compliance.

Input: $ARGUMENTS

Usage

/audit                    # Full infrastructure audit
/audit --framework        # Focus on framework setup only
/audit --tools           # Focus on tool installation
/audit --fix             # Apply fixes for detected issues

Phase 1: Framework Setup Validation

Directory Structure Audit

Verify .claude/ directory exists and has proper structure
Check for required subdirectories: work/, memory/, reference/
Validate permissions on framework directories
Ensure proper gitignore entries for sensitive files

Configuration Validation

Check settings.json exists and has valid syntax
Validate hook configurations if present
Verify tool permissions are properly configured
Ensure CLAUDE.md hierarchy is properly structured

Memory System Health

Verify memory files are not corrupted
Check import links are valid (no broken @references)
Validate memory file sizes are within limits
Ensure session memory is properly rotated

Phase 2: Tool Installation Audit

Core Dependencies

Git: Version check, configuration validation
Python Tools: ruff, black, mypy, pytest availability
JavaScript Tools: eslint, prettier, jest (if applicable)
System Tools: jq, flock, timeout, mktemp

Git Configuration

Verify git safe-commit alias is configured
Check user name and email are set
Validate pre-commit hooks are installed
Ensure conventional commit format is enforced

Language-Specific Tools

Python Projects: Check pyproject.toml, requirements.txt
JavaScript Projects: Validate package.json, node_modules
Go Projects: Verify go.mod, tool installations
Rust Projects: Check Cargo.toml, rust toolchain

Phase 3: Quality Infrastructure Audit

Code Quality Tools

Verify linting tools are properly configured
Check formatting tools work correctly
Validate type checking is enabled where appropriate
Ensure test runners are properly set up

Read the full file on GitHub · 175 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.

yesterday First seen · 175 lines · 6 tokens per session scan A 16f51388a4f3

Subscribe to this mod's changes

audit is a command published in the GitHub repository stefan-jansen/claude-code-toolkit (85 stars, last pushed 1mo ago), licensed MIT. It adds 6 tokens to every session and 1,193 once invoked, about $0.0000 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.

Memory and context agents anthropic claude-code coding-agents framework plugins

audit