mcpField
M
MCPAmpel - MCP Security Scanner

MCPAmpel - MCP Security Scanner

Local
github.com/MCPAmpelSecurity2Updated 30w ago
Repository

About MCPAmpel - MCP Security Scanner

MCPAmpel is a security scanner for installed MCP servers, aimed at developers using AI agents. It discovers servers from several editor configurations and aggregates results and trust scores from 16 scanning engines.

Key features

  1. 01Discover MCP servers in editor configurations
  2. 02Scan servers with 16 engines
  3. 03Aggregate trust scores and detailed findings
  4. 04Scan GitHub, GitLab, npm, or PyPI URLs
  5. 05Check daily quota and scan results

Use cases

  1. 01Review risks in configured MCP servers
  2. 02Scan a project URL before connecting to it
  3. 03Inspect findings and check daily usage

MCP support

MCPAmpel is an MCP server that provides tools to discover and scan servers, scan a single URL, check quota status, and retrieve scan results.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "diemoeve-mcpampel": {
      "command": "uvx",
      "args": [
        "mcpampel"
      ],
      "env": {
        "MCPAMPEL_API_KEY": "<MCPAMPEL_API_KEY>"
      }
    }
  }
}
Run commanduvx mcpampel
Environment variables to set
MCPAMPEL_API_KEYSecretMCPAmpel API key (get one free at mcpampel.com)

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypimcpampel0.2.2

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

MCPAmpel - MCP Security Scanner

MCPAmpel

Scan your installed MCP servers for security vulnerabilities, directly from your AI agent.

MCPAmpel discovers MCP servers from your Claude Code, Cursor, Windsurf, or Gemini CLI configuration, submits them to 16 scanning engines, and returns an aggregated trust score with detailed findings.

50 API calls/day included.

Quick Start

uvx mcpampel

Configuration

Claude Code / Claude Desktop

Add to ~/.claude/settings.json or claude_desktop_config.json:

{
  "mcpServers": {
    "mcpampel": {
      "command": "uvx",
      "args": ["mcpampel"],
      "env": {
        "MCPAMPEL_API_KEY": "your_key_here"
      }
    }
  }
}
Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "mcpampel": {
      "command": "uvx",
      "args": ["mcpampel"],
      "env": {
        "MCPAMPEL_API_KEY": "your_key_here"
      }
    }
  }
}
Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "mcpampel": {
      "command": "uvx",
      "args": ["mcpampel"],
      "env": {
        "MCPAMPEL_API_KEY": "your_key_here"
      }
    }
  }
}
Gemini CLI

Add to ~/.gemini/settings.json:

{
  "mcpServers": {
    "mcpampel": {
      "command": "uvx",
      "args": ["mcpampel"],
      "env": {
        "MCPAMPEL_API_KEY": "your_key_here"
      }
    }
  }
}
Getting an API Key

Register for free at mcpampel.com.

Environment Variables
VariableRequiredDefaultDescription
MCPAMPEL_API_KEYYes-Your API key (free at mcpampel.com)
MCPAMPEL_BASE_URLNohttps://mcpampel.comAPI base URL

Tools

scan_my_servers

Discovers all MCP servers from your editor config and scans them with 16 engines. Returns a summary table with trust scores. No input needed.

scan_url

Scan a single GitHub, GitLab, npm, or PyPI URL. Returns trust score, engine breakdown, and findings.

ParameterTypeRequired
urlstringYes
check_status

Show your daily quota usage and remaining scans.

get_scan_results

Get detailed results for a specific scan by ID. Use after scan_my_servers or scan_url to drill into findings.

ParameterTypeRequired
scan_idstringYes

Development

cd mcp-plugin
uv sync
uv run pytest

License

Apache License 2.0

mcpampel.com

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.