This post will focus on an alternative method of using Kali Linux, moving beyond direct terminal command execution. Instead, we will leverage a Large Language Model (LLM) to translate “natural language” descriptions of desired actions into technical commands. Achieving this setup requires the integration of three distinct systems:

Kali Linux

A step-by-step guide to integrating Kali Linux with Claude Desktop on macOS via Model Context Protocol (MCP), enabling natural language control of penetration testing tools. The setup involves three components: macOS running Claude Desktop as the UI and MCP client, a Kali Linux instance (local or cloud) as the attack box, and Anthropic's Claude Sonnet 4.5 as the LLM. The guide covers SSH key setup between macOS and Kali, installing mcp-kali-server along with tools like nmap, gobuster, nikto, and metasploit, and configuring Claude Desktop's MCP client to SSH into Kali and execute commands. The result allows users to issue natural language prompts like 'port scan scanme.nmap.org' and have the LLM translate and execute the appropriate commands automatically.

Kali & LLM: macOS with Claude Desktop GUI & Anthropic Sonnet LLM