home / mcp / exegol mcp server

Exegol MCP Server

Provides an MCP server to control Exegol pentest containers via predefined workflows and command execution.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "exho7-exegol-mcp": {
      "command": "python",
      "args": [
        "/absolute/path/to/exegol_mcp.py"
      ]
    }
  }
}

You deploy the Exegol MCP Server to let AI agents run predefined pentesting workflows inside Exegol containers. It provides a stable MCP interface to execute commands, list containers, check health, and run complex workflows with automatic sequencing and error handling.

How to use

To use the Exegol MCP Server, start the server as a local process and connect a client that supports MCP over stdio. You will be able to list available Exegol containers, execute commands inside a container, check server health, and run one of the predefined pentesting workflows. Use the client to select a workflow, configure required parameters, and let the server orchestrate the sequence of steps across containers.

How to install

Prerequisites you need before running the server are installed Python 3.10 or newer, Docker, and the Exegol CLI, with at least one Exegol container created.

Install the Python dependencies for the server by installing from a requirements file.

Additional sections

Configuration, security notes, and practical usage details follow. Below are practical steps to set up and run the server, integrate with clients, and understand available tools and workflows.

exegol:
  path: "exegol"  # or /usr/local/bin/exegol

timeout:
  command_execution: 600  # 10 minutes

logging:
  level: "INFO"  # DEBUG, INFO, WARNING, ERROR
  format: "json"  # json or text

mcp:
  server_name: "exegol-mcp-server"
  version: "0.1.0"
  compact_mode: true

sessions:
  enabled: false
  idle_timeout: 300

parsing:
  auto_parse: true

MCP server connection and config (summary)

The server runs as a local, stdio-based process. You start it with Python and point to the server script. An example command sequence shown for integration is to run the server script directly and then connect a client that communicates over stdio.

Troubleshooting and notes

If you encounter issues starting the server, verify that Python is 3.10+ and that Docker is running. Ensure the Exegol CLI can list containers and that at least one container exists. If a workflow fails, check whether steps can continue on failure and inspect the per-step outputs for debugging.

Security and usage tips

Only run the MCP Server in trusted environments. Use the compact mode to reduce token usage in AI responses and enable auto-parse to structure tool outputs for easier analysis. Consider enabling persistent sessions only if you need sequential commands across multiple steps on the same container.

Examples and workflows

You can run predefined workflows such as web reconnaissance, port scanning, and vulnerability assessment. The server exposes tools to list containers, execute commands, check status, and run workflows with automatic step sequencing and error handling.

Available tools

exegol_exec

Execute a command inside an Exegol container and return stdout, stderr, and exit status.

exegol_list

List all available Exegol containers.

exegol_status

Check MCP server health status and current load.

list_workflows

List all predefined pentesting workflows with their metadata.

run_workflow

Execute a selected pentesting workflow with required parameters.