Claudit Sec
Security audit tool for Claude Desktop and Claude Code on macOS — single-command visibility into MCP servers, extensions, plugins, connectors, scheduled tasks, and permissions.
What is Claudit Sec?
Claudit Sec is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to security audit tool for claude desktop and claude code on macos — single-command visibility into mcp servers, extensions, plugins, connectors, scheduled tasks, and permissions.
Security audit tool for Claude Desktop and Claude Code on macOS — single-command visibility into MCP servers, extensions, plugins, connectors, scheduled tasks, and permissions.
This server falls under the Security category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Security audit tool for Claude Desktop and Claude Code on ma
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx claudit-secConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Claudit Sec
claudit-sec is a read-only security audit tool for Claude Desktop and Claude Code on macOS and Windows that provides single-command visibility into the full attack surface of your Claude installation. It inspects MCP server registrations, browser-control extensions with OAuth credentials, plugin lifecycle hooks that execute shell commands, scheduled tasks with cron expressions, and agent-to-agent connectors — then outputs a human-readable terminal report, an HTML summary, or structured JSON for SIEM ingestion. Security teams and privacy-conscious developers use it to verify that no unauthorized or unexpected integrations have been added to their Claude environment.
Prerequisites
- Claude Desktop or Claude Code installed on macOS or Windows
- macOS: zsh (built-in since Catalina) and jq installed (brew install jq)
- Windows: PowerShell 5.1 or higher (ships with Windows 10+)
- git to clone the repository (no npm or pip dependencies required)
- Read access to Claude configuration directories (no elevated privileges needed)
Clone the claudit-sec repository
Clone the repository to your local machine. No package manager installation is needed — the tool is a self-contained shell script.
git clone https://github.com/HarmonicSecurity/claudit-sec.git
cd claudit-secInstall jq on macOS (if not present)
The macOS audit script requires jq to parse JSON configuration files. Install it via Homebrew if not already available.
brew install jq
# Verify
jq --versionMake the script executable and run the audit (macOS)
Grant execute permission to the audit script and run it. The script is read-only and never modifies or deletes any audited file.
chmod +x claude_audit.sh
./claude_audit.shRun the audit on Windows
On Windows, run the PowerShell audit script with the bypass execution policy flag. No additional dependencies are required.
powershell -ExecutionPolicy Bypass -File claude_audit.ps1Generate output in different formats
Choose the output format that fits your workflow: terminal text (default), HTML report for sharing, or JSON for SIEM or automated processing.
# HTML report
./claude_audit.sh --html
# JSON for SIEM
./claude_audit.sh --json > audit.json
# Warnings only (quiet mode)
./claude_audit.sh -qClaudit Sec Examples
Client configuration
claudit-sec is a standalone audit CLI and does not run as an MCP server itself. It audits MCP server configs already present in Claude Desktop. The typical usage is running the script directly, not via a config block. If you want to invoke it from within Claude Code as a tool, you can register it as a custom command.
{
"mcpServers": {
"claudit-sec": {
"command": "bash",
"args": ["/path/to/claudit-sec/claude_audit.sh", "--json"],
"env": {}
}
}
}Prompts to try
Audit scenarios and checks you can run with claudit-sec
- Run ./claude_audit.sh to get a full terminal report of all MCP servers and hooks
- Run ./claude_audit.sh --html to generate a shareable HTML audit report
- Run ./claude_audit.sh --json | jq '.mcpServers' to extract just MCP server entries
- Run ./claude_audit.sh -q to see only warnings and high-severity findings
- Run ./claude_audit.sh --json > baseline.json to capture a baseline for comparisonTroubleshooting Claudit Sec
jq: command not found on macOS
Install jq with brew install jq. If Homebrew is not installed, visit https://brew.sh for the one-liner installation command. After installing jq, re-run the audit script.
PowerShell script blocked by execution policy on Windows
Run the script with the bypass flag: powershell -ExecutionPolicy Bypass -File claude_audit.ps1. This only affects the current session and does not permanently change system policy.
Audit reports 'no Claude config found' or empty output
Confirm Claude Desktop or Claude Code is installed and has been run at least once so config files exist. On macOS, configs are typically at ~/Library/Application Support/Claude/. The script searches these default locations automatically.
Frequently Asked Questions about Claudit Sec
What is Claudit Sec?
Claudit Sec is a Model Context Protocol (MCP) server that security audit tool for claude desktop and claude code on macos — single-command visibility into mcp servers, extensions, plugins, connectors, scheduled tasks, and permissions. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Claudit Sec?
Follow the installation instructions on the Claudit Sec GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Claudit Sec?
Claudit Sec works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Claudit Sec free to use?
Yes, Claudit Sec is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Claudit Sec Alternatives — Similar Security Servers
Looking for alternatives to Claudit Sec? Here are other popular security servers you can use with Claude, Cursor, and VS Code.
Casdoor
★ 13.6kAn open-source Agent-first Identity and Access Management (IAM) /LLM MCP & agent gateway and auth server with web UI supporting OpenClaw, MCP, OAuth, OIDC, SAML, CAS, LDAP, SCIM, WebAuthn, TOTP, MFA, Face ID, Google Workspace, Azure AD
ghidraMCP
★ 9.0kAn Model Context Protocol server that enables LLMs to autonomously reverse engineer applications by exposing Ghidra's decompilation and analysis tools. It allows AI agents to list code structures, rename methods, and analyze binaries directly through
HexStrike AI
★ 8.9kHexStrike AI MCP Agents is an advanced MCP server that lets AI agents (Claude, GPT, Copilot, etc.) autonomously run 150+ cybersecurity tools for automated pentesting, vulnerability discovery, bug bounty automation, and security research. Seamlessly b
IDA Pro MCP
★ 8.7kEnables AI-assisted reverse engineering in IDA Pro by providing tools to analyze binaries, decompile functions, manage comments, search patterns, and interact with the IDA database through natural language.
Anthropic Cybersecurity Skills
★ 6.6k754 structured cybersecurity skills for AI agents · Mapped to 5 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platform
Hooker
★ 5.1k🔥🔥 hooker is a Frida-based reverse engineering toolkit for Android. It offers a user-friendly CLI, universal scripts, auto hook generation, memory roaming to detect activities/services, one-click SOCKS5 proxy setup, Frida JustTrustMe, and BoringSSL u
Browse More Security MCP Servers
Explore all security servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Claudit Sec in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Claudit Sec?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.