MCPSentinel

Overview

MCPSentinel is a Model Context Protocol server that delivers precision-first security scanning for MCP servers. Once integrated into your AI assistant, it exposes security scanning capabilities as tools your model can invoke automatically during development and deployment.

The server runs locally as a Python package via uvx mcp-guardian-scan, communicating over stdio. You'll need to set the MCPSENTINEL_ALLOWED_HOSTS environment variable to configure which hosts the scanner can target. Version v0.8.7 is available under the MIT license, with full source code at gentaArnezzi/MCPSentinel on GitHub.

MCPSentinel appears in the Version Control category and works with MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, and Zed. If you maintain MCPSentinel, you can claim this listing to verify ownership and keep the information up to date.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "mcpsentinel": {
      "command": "uvx",
      "args": [
        "mcp-guardian-scan"
      ],
      "env": {
        "MCPSENTINEL_ALLOWED_HOSTS": "<mcpsentinel-allowed-hosts>"
      }
    }
  }
}

Requires MCPSENTINEL_ALLOWED_HOSTS