MCP Sentinel

Security scanner for Model Context Protocol (MCP) servers. Detects vulnerabilities before attackers do.

Installation

# Run directly with npx (recommended)
npx @zcode-apps/mcp-sentinel scan https://your-mcp-server.com

# Or install globally
npm install -g @zcode-apps/mcp-sentinel
mcp-sentinel scan https://your-mcp-server.com

Usage

# Basic scan (text output)
npx @zcode-apps/mcp-sentinel scan https://api.example.com/mcp

# JSON output
npx @zcode-apps/mcp-sentinel scan https://api.example.com/mcp --json

# Verbose mode (show evidence)
npx @zcode-apps/mcp-sentinel scan https://api.example.com/mcp --verbose

Features

• MCP Protocol Detection - Verifies valid MCP endpoints
• Authentication Bypass - Checks for missing auth
• Dangerous Tool Detection - Finds tools with RCE, file access, SQL risks
• Path Traversal - Detects unsafe resource URIs
• Prompt Injection Risks - Identifies dynamic prompt vulnerabilities

Output Example

🔍 MCP Sentinel - Security Scanner

Target: https://api.example.com/mcp
──────────────────────────────────────────────────

🔵 INFO:

   [INFO]
   MCP Server detected: my-mcp-server v1.0.0
   
   [INFO]
   Found 5 exposed tools: ["get_weather", "run_command", "read_file"]

🟠 HIGH SEVERITY:

   [AUTH_BYPASS]
   MCP server accepts unauthenticated connections
   💡 Recommendation: Implement authentication on MCP endpoints

──────────────────────────────────────────────────
📊 SUMMARY:
   Critical: 0
   High:     1
   Medium:   0
   Info:     2

Why MCP Sentinel?

| Stat | Value | |------|-------| | MCP servers vulnerable to RCE | 43% | | Exposed MCP servers worldwide | 5,200+ | | Documented CVEs | 60+ |

Don't be part of the 43%. Scan your MCP servers today.

Repository

GitHub: https://github.com/zcode-apps/mcp-sentinel
npm: https://www.npmjs.com/package/@zcode-apps/mcp-sentinel

License

MIT License

Built by Sebastian Zang