MCP Watch
** - A comprehensive security scanner for Model Context Protocol (MCP) servers that detects vulnerabilities and security issues in your MCP server implementations.
What is MCP Watch?
MCP Watch is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ** - a comprehensive security scanner for model context protocol (mcp) servers that detects vulnerabilities and security issues in your mcp server implementations.
** - A comprehensive security scanner for Model Context Protocol (MCP) servers that detects vulnerabilities and security issues in your MCP server implementations.
This server falls under the Security category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- ** - A comprehensive security scanner for Model Context Prot
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx mcp-watchConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use MCP Watch
MCP Watch is a security scanner specifically designed for Model Context Protocol server implementations. It analyzes MCP server code to detect 12 categories of vulnerabilities including credential leakage, tool poisoning, prompt injection, parameter injection, conversation exfiltration, ANSI injection, and permission misconfigurations. Security engineers and MCP developers use it to audit their server implementations before deployment, either scanning remote GitHub repositories or local project directories.
Prerequisites
- Node.js and npm installed
- Git installed (for cloning the source if building from scratch)
- Optional: Docker for running the containerized version
- The MCP server source code you want to scan (local path or GitHub URL)
Install mcp-watch globally
Install the mcp-watch CLI tool globally via npm so it is available from any directory.
npm install -g mcp-watchScan a GitHub repository
Point mcp-watch at a public GitHub repository containing an MCP server implementation. Results are printed to the console by default.
mcp-watch scan https://github.com/owner/mcp-server-repoScan a local project directory
Run the scanner against a local directory. Use `.` to scan the current working directory.
mcp-watch scan-local /path/to/your/mcp-projectFilter by severity and output format
Filter results to show only high or critical findings, and output as JSON for integration with other tools.
mcp-watch scan https://github.com/owner/mcp-server --format json --severity highFilter by vulnerability category
Scan only for a specific category of vulnerability, such as credential detection or prompt injection.
mcp-watch scan-local . --category prompt-injectionMCP Watch Examples
Scanning an MCP server and capturing JSON output
Run a high-severity scan against a local MCP project and save the results as JSON for further processing.
{
"command": "mcp-watch",
"args": ["scan-local", ".", "--format", "json", "--severity", "high"],
"description": "Scan the current directory for high/critical MCP vulnerabilities"
}CLI commands to try
Common mcp-watch CLI invocations for different scanning scenarios.
- "mcp-watch scan https://github.com/owner/my-mcp-server" — full scan of a GitHub repo
- "mcp-watch scan-local . --severity critical" — only show critical findings in current dir
- "mcp-watch scan-local ./src --format json > report.json" — save JSON report
- "mcp-watch scan https://github.com/owner/server --category credential-detection" — check for leaked credentials onlyTroubleshooting MCP Watch
mcp-watch scan fails for a private GitHub repository
mcp-watch requires public repository access when scanning GitHub URLs. For private repositories, clone the repo locally first and use `mcp-watch scan-local /path/to/clone`.
No vulnerabilities reported but the code clearly has issues
Try lowering the severity threshold: `--severity low` will surface all findings including informational ones. Also check that the code files are in a supported language (JavaScript/TypeScript are the primary targets).
npm install -g mcp-watch fails with permission errors
Use `sudo npm install -g mcp-watch` on Linux/macOS, or configure npm to use a user-writable global directory with `npm config set prefix ~/.npm-global` and add `~/.npm-global/bin` to your PATH.
Frequently Asked Questions about MCP Watch
What is MCP Watch?
MCP Watch is a Model Context Protocol (MCP) server that ** - a comprehensive security scanner for model context protocol (mcp) servers that detects vulnerabilities and security issues in your mcp server implementations. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install MCP Watch?
Follow the installation instructions on the MCP Watch GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with MCP Watch?
MCP Watch works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is MCP Watch free to use?
Yes, MCP Watch is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
MCP Watch Alternatives — Similar Security Servers
Looking for alternatives to MCP Watch? 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 MCP Watch 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 MCP Watch?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.