Enables execution of curl commands for HTTP requests and web application interaction as part of penetration testing and CTF challenge solving workflows
References GitHub repository hosting and mentions the 5ire MCP client available on GitHub for connecting to the Kali terminal bridge
Contact method for the project author
Provides API bridge to execute Kali Linux terminal commands remotely, enabling AI agents to run penetration testing tools like nmap, gobuster, nuclei, and other security tools for automated reconnaissance and vulnerability assessment
Supports integration with OpenAI's API endpoints for AI-assisted penetration testing and security analysis workflows
Supports integration with Signal for secure alerting and notification pipelines to report security findings
Supports integration with Slack for alerting and notification pipelines to report security findings and scan results
Can store scan results and evidence in SQLite databases for RAG and state management in AI-driven security workflows
MCP Kali Server
Kali MCP Server — by Lakshmikanthan (Letchu) — GitHub: letchupkt
Lightweight API bridge that connects MCP Clients (examples: Claude Desktop, 5ire) to a Kali Linux terminal. It enables MCPs to execute terminal commands, interact with web applications, and run AI-assisted offensive security workflows.
🔍 Overview
Kali MCP Server exposes a controlled API that allows authorized MCP clients to remotely run commands (e.g. nmap, curl, gobuster) and receive structured output. The project is designed to accelerate automated penetration testing, CTF solving, and AI-driven reconnaissance by combining Kali tooling with local or remote LLMs.
Use-cases include:
AI-assisted penetration testing and bug hunting.
Solving web CTF challenges in real time (useful for HTB/THM/CTF practice).
Orchestrating recon/exploit chains through a single MCP interface.
🚀 Key Features
🧠 AI Endpoint Integration — connect the Kali host to any MCP or LLM endpoint (OpenAI, Claude, DeepSeek, or other local models).
🖥️ Command Execution API — run terminal tools remotely and get structured responses.
🕸️ Web Challenge & CTF Support — AI agents can interact with target web apps (via
curl,wget, fuzzers) and attempt to capture flags or findings.🔐 Built for Offensive Security Practitioners — red teamers, bug bounty hunters and CTF players in mind.
🎯 Large Toolset — ships with (or can install) 25+ specialized tools for reconnaissance, scanning, and exploitation.
🛠️ Integrated Tools
Recon & Subdomain Discovery: subfinder, sublist3r, amass, assetfinder
Subdomain Takeover: subzy, subjack
HTTP Probing & Crawling: httpx, katana, gospider
Content / Dir Discovery: gobuster, dirb, ffuf, feroxbuster, dirsearch
Vuln Scanners: nuclei, nikto, wpscan, dalfox
Param / URL Discovery: arjun, paramspider, waybackurls, gau, gf
Port Scanners: nmap, masscan, rustscan
Auth / Passwords: hydra, john, sqlmap
Network Enumeration: enum4linux
This list is configurable — tools can be added/removed depending on your Kali environment and permitted scope.
🛠️ Quick Install (Kali host)
Note: install_tools.sh will try to install many common pentest binaries. Inspect it and run it inside a VM or disposable Kali box if you prefer.
🖥️ Client Configuration (MCP Clients)
Clients simply invoke the MCP script with the Kali host URL. Example command used by MCP clients:
Claude Desktop
Edit %APPDATA%\Claude\claude_desktop_config.json and add an MCP entry:
5ire Desktop
Add an MCP using the same python3 /absolute/path/to/mcp_server.py http://LINUX_IP:5000 command — 5ire will generate the needed configuration automatically.
🧩 Example Workflows
CTF Web Challenge (high-level):
Client sends challenge URL + scope to MCP.
MCP runs
httpx/gobuster/ffufand returns parsed results.Model suggests follow-up fuzzing or
sqlmapcommands.Operator reviews and executes exploit steps, MCP logs outputs.
HTB-like Machine (non-destructive recon):
Run
nmap/rustscanthenenum4linuxfor SMB.Save and store findings in local vector DB for the AI to reason over.
⚙️ Suggested Architecture & Safety
Orchestration layer: Use a small Python service to call system tools and normalize outputs to JSON.
RAG & State: Store scan results and evidence in a local DB (e.g., SQLite + FAISS) and pass concise context to the LLM.
Sandboxing: Run scans/exploits inside VMs/containers with strict network egress control.
Authorization checks: Add a required authorization token and a pre-flight scope/consent checklist for every automated run.
🔒 Legal & Ethical Notice
Only run this server against systems you own or explicitly have written authorization to test. Automated pentesting can cause service disruption and legal consequences. This repository is intended for education, defensive research, and authorized testing only.
🔮 Other Possibilities
Memory forensics with Volatility (automated workflows).
Disk forensics (SleuthKit / timelines / carving).
Integration with alerting pipelines (Slack, Signal, email) for findings.
Local LLM orchestration (run quantized LLMs for on-device reasoning — tell me your hardware and I can recommend exact models and setup commands).
Contributing
PRs welcome. If you add tools or workflows, include tests and updated docs. Please keep contributions scoped to legal, ethical tooling.
Author & Contact
Lakshmikanthan (Letchu)
GitHub: letchupkt
Portfolio: https://letchupkt.vgrow.tech
Instagram: @letchu_pkt
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.
Enables AI assistants to execute penetration testing commands and security tools on Kali Linux remotely. Supports automated reconnaissance, vulnerability scanning, and CTF solving through integration with 25+ offensive security tools like nmap, gobuster, and nuclei.