Home / Security tooling

agent-config-audit

Audit AI agent configuration files for security risks.

Python · MIT · latest v0.3.0

What it is

One command reads the files that launch and instruct a coding agent (.claude/settings*.json, .mcp.json, claude_desktop_config.json, .cursor/ rules and MCP config, CLAUDE.md and the files it imports, AGENTS.md, plugin manifests, hooks, skills) and reports pre-approved dangerous commands, bypassed permission prompts, rules Claude Code silently skips, secrets next to server definitions, unpinned MCP servers, hooks that phone home and prompt-injection text. Output as a table, JSON, Markdown or SARIF for GitHub code scanning. Read-only, standard library only, no network.

Who it is for: developers who open other people's repositories with Claude Code, Cursor or another agent, and teams who review .claude/ and .mcp.json changes in pull requests.

Install

pipx install git+https://github.com/basitalisandhu/agent-config-audit
cd your-project
agent-config-audit --fail-on high

From the README; see the full README for every option.

Releases

More in Security tooling

All tools

Topics: agent-security, ai-agents, ai-security, claude-code, claude-md, cli, configuration-audit, cursor, devsecops, github-actions, hacktoberfest, llm-security, mcp, pre-commit, prompt-injection, python, sarif, secrets-detection, security-automation, security-tools