ITMCP
Secure network administration tools for AI assistants through the Model Context Protocol (MCP).
Overview
ITMCP is an MCP server that enables AI assistants to safely execute networking commands inside a Docker container sandbox. It provides a secure interface for running common network diagnostic and administration tools while maintaining strict security controls.
The project implements the Model Context Protocol (MCP) to expose networking tools as callable functions for AI assistants, allowing them to perform network diagnostics and system administration tasks in a controlled environment.
Features
- Docker Isolation: All commands run in a sandboxed Docker container for enhanced security
- Security Controls: Comprehensive whitelisting of hosts, directories, and commands
- Network Diagnostic Tools: SSH, ping, nslookup, telnet, dig, tcpdump, and more
- File Operations: Secure access to view and analyze files with tools like cat, grep, head, tail
- Process Management: View running processes with ps and top tools
- Credential Management: Secure handling of SSH keys and passwords
- MCP Integration: Full compatibility with the Model Context Protocol
- Enterprise-Grade Security: Session management, audit logging, and access controls
Installation
Prerequisites
- Python 3.10 or higher
- Docker (for containerized execution)
- MCP library (version 1.0.0 or higher)
Basic Installation
- Clone the repository:
- Install dependencies:
Docker Setup
- Build the Docker container:
- Run the container:
Configuration
ITMCP uses a YAML-based configuration system and environment variables for setup.
Environment Variables
Create a .env
file in the project root with the following variables:
Security Whitelists
ITMCP implements three key whitelists for security:
- Allowed Hosts: Restricts which hosts can be targeted by network tools
- Allowed Directories: Limits file system access to specific directories
- Allowed Remote Commands: Controls which commands can be executed remotely
Available Tools
ITMCP provides the following network administration tools:
Tool | Description |
---|---|
ssh_tool | Connect to a target via SSH |
ping_tool | Ping a host to check connectivity |
nslookup_tool | Perform DNS lookup on a hostname or IP address |
telnet_tool | Test TCP connectivity to a host and port |
dig_tool | Perform DNS lookup with dig command |
tcpdump_tool | Capture network packets (limited time) |
ps_tool | List running processes |
cat_tool | Display content of a file |
top_tool | Display system processes (snapshot) |
grep_tool | Search for patterns in files |
head_tool | Display the beginning of a file |
tail_tool | Display the end of a file |
Security Features
ITMCP implements enterprise-grade security features:
Session Management
- Secure session creation with cryptographic tokens
- Session expiration and timeout controls
- Concurrent session limits
- Session validation and regeneration
Audit Logging
- Comprehensive command logging
- User attribution for all actions
- Success/failure logging
- Security event flagging
- Tamper-evident logs
Access Control
- Command whitelisting
- Directory restrictions
- Host restrictions
- Input validation and sanitization
Docker Integration
ITMCP uses Docker to create a secure sandbox for command execution:
- All commands are routed through the Docker container
- The container has limited access to the host system
- Resource limits can be applied to prevent abuse
- Network isolation provides an additional security layer
Usage Examples
MCP Configuration
Claude Desktop Configuration
To use ITMCP with Claude desktop, add the following to your config.json
file:
Cline Configuration
For Cline AI, a more detailed configuration is provided in the mcp-config.json
file included in this repository:
To use this configuration with Cline:
- Copy the
mcp-config.json
file to your Cline configuration directory - Start Cline with the
--mcp-config
flag pointing to this file - The ITMCP tools will be available for use in your Cline sessions
Example 1: Ping a Host
Example 2: SSH Connection to Firewall
Example 3: DNS Lookup
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
- Fork the repository
- Create your feature branch (
git checkout -b feature/amazing-feature
) - Commit your changes (
git commit -m 'Add some amazing feature'
) - Push to the branch (
git push origin feature/amazing-feature
) - Open a Pull Request
License
This project is licensed under the MIT License - see the LICENSE file for details.
Author
Andrew Hopper
- Email: hopperab@gmail.com
- Twitter: x.com/andrewhopper
- Website: andyhop.316.dev
- LinkedIn: linkedin.com/in/andrewhopper
Security Considerations
ITMCP is designed with security in mind, but proper configuration is essential:
- Always run in a Docker container for isolation
- Carefully configure whitelists for hosts, directories, and commands
- Regularly review audit logs for suspicious activity
- Keep the system updated with security patches
- Follow the security best practices in the documentation
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
Secure network administration MCP server that enables AI assistants to safely execute networking commands like SSH, ping, and DNS lookups inside a Docker container sandbox.
Related MCP Servers
- -securityAlicense-qualityAn MCP server to create secure code sandbox environment for executing code within Docker containers.Last updated -69GoMIT License
- -securityAlicense-qualityA secure, container-based implementation of the Model Context Protocol (MCP) that provides sandboxed environments for AI systems to safely execute code, run commands, access files, and perform web operations.Last updated -4PythonApache 2.0
- -securityAlicense-qualityAn enhanced MCP server that grants AI assistants the ability to execute terminal commands on a user's system with improved security controls, designed for use in controlled environments.Last updated -PythonMIT License
- -securityFlicense-qualityA secure MCP server that provides controlled ShellJS access for LLMs, enabling AI systems to safely execute shell commands and interact with the filesystem within a configurable security sandbox.Last updated -JavaScript