REDCLAW
AI
[ AUTONOMOUS OFFENSIVE SECURITY AGENT ]
Deploy an AI agent that autonomously plans and executes red team operations on your machine. 28+ offensive tools, multi-provider LLM routing, and full computer control — self-hosted, open source, built for professionals.
Everything a Red Teamer Needs
A complete autonomous agent stack — from initial recon to final report.
Autonomous Agent Loop
Think → Plan → Execute → Observe. RedClaw AI autonomously breaks down red team objectives into multi-step operations, selects the right tools, interprets results, and adapts its strategy in real time — no hand-holding required.
28+ Offensive Tools
The full offensive security stack: nmap, masscan, nuclei, sqlmap, ffuf, metasploit, impacket, hashcat, hydra, and 19+ more. Missing a tool? The agent detects it and installs it automatically.
Multi-Provider LLM Routing
Route each task to the optimal model. Brain tasks go to GPT-4.1 or Claude Sonnet. Worker tasks use fast cheap models like Groq Llama. Coder tasks hit DeepSeek. Analyst reports use GPT-4.1-mini. Minimize cost, maximize performance.
Full Browser Control
Playwright-powered headless browser automation. Navigate web targets, interact with forms, extract cookies, take screenshots, and test web application logic — all orchestrated by the agent.
Network & Recon Stack
Nmap scanning profiles, subdomain enumeration, DNS lookups, SSL certificate analysis, port scanning, WHOIS lookups, and HTTP header analysis — all wired directly into the agent's tool belt.
Automated Pentest Reports
Generate professional penetration testing reports in Markdown with severity ratings, CVSS scores, evidence, impact analysis, and remediation recommendations. Shareable, structured, and ready for clients.
28+ Offensive Security Tools
All auto-installed on demand. The agent handles dependencies — you focus on the mission.
Right Model,
Right Task
Stop paying GPT-4 prices for simple shell commands. RedClaw AI routes each task to the most cost-effective model for that role — keeping expensive tokens for the work that actually needs them.
From Zero to Exploitation
One-command setup on Kali Linux, Docker, or your local machine. The agent is running in under 5 minutes.
Set your LLM provider API keys and configure role routing. Mix providers to minimize cost and maximize capability.
Submit a red team objective via the web dashboard or REST API. Natural language — no scripting required.
The agent plans, selects tools, executes operations, interprets results, and delivers a structured report.
Deploy Anywhere
Four deployment paths. All self-hosted. All yours.
Full native install with systemd service. Best for dedicated red team VMs.
Lightweight containerized deployment. Ready in under 2 minutes.
Full Kali toolset inside Docker. All 40+ tools pre-installed.
Quick setup on any Linux/macOS machine. Zero configuration.
Ready to Deploy?
Open source, self-hosted, and built for professionals. No subscriptions, no data leaving your machine.
Built in the Open. Trusted by Red Teamers.
MIT License — fork it, extend it, own it.
Zero telemetry. Your keys, your data, your machine.
Built for professional security testing with explicit permission.