██████╗ ██████╗ ██╗ ██╗ █████╗ ██████╗ ██╗
██╔══██╗██╔══██╗██║ ██╔╝██╔══██╗██╔════╝ ██║
██║ ██║██████╔╝█████╔╝ ███████║██║ ███╗██║
██║ ██║██╔══██╗██╔═██╗ ██╔══██║██║ ██║██║
██████╔╝██║ ██║██║ ██╗██║ ██║╚██████╔╝██║
╚═════╝ ╚═╝ ╚═╝╚═╝ ╚═╝╚═╝ ╚═╝ ╚═════╝ ╚═╝
The AI-powered penetration testing framework built for elite Red Teams.
DRKagi combines 80+ security tools with AI reasoning to automate reconnaissance, exploitation, and reporting. It thinks like a pentester — adapting attacks, evading firewalls, mapping to MITRE ATT&CK, and generating professional reports.
| Feature | Description |
|---|---|
| 🧠 AI Brain | Groq Llama 3.3 70B — reasons about findings and chains attacks |
| 🛡️ 80+ Tools | Nmap, SQLMap, Hydra, Metasploit, Nuclei, BloodHound, and more |
| 👻 Stealth by Default | 6 firewall evasion techniques applied automatically |
| 🔑 Multi-Key API | Rotate up to 50 Groq keys — auto-cooldown on rate limits |
| 🤖 Autopilot | 4-phase automated assessment (single target or entire subnet) |
| 🎭 AI Personas | 5 modes: Ghost, Blitz, CTF, Recon, Web Hunter |
| 🧩 Plugin System | Drop .py scripts in plugins/ to add custom commands |
| 🗂️ Profiles & Sessions | Save/resume engagements across restarts |
| 🔐 Credential Vault | AES-encrypted storage for discovered credentials |
| 🗺️ MITRE ATT&CK | Every suggestion mapped to ATT&CK technique IDs |
| 💭 Chain-of-Thought | See the AI's reasoning process step by step |
| 📜 Script Generator | AI writes custom Python/Node.js tools on demand |
| 📊 Dashboard | Streamlit web UI with network topology maps |
| 📄 PDF Reports | AI-written executive summaries + MITRE mapping |
| ✅ Compliance | Map findings to PCI-DSS, HIPAA, ISO 27001 |
| 🌐 REST API | Flask API for Burp/ZAP/tool integration |
| 🐳 Docker | One command: docker run drkagi |
curl -sL https://raw.githubusercontent.com/DRKagi/DRKagi/main/install.sh | bashcd ~/DRKagi && git pull origin main && python3 -m venv .venv && source .venv/bin/activate && pip install -r requirements.txt -q && drkagi
cd ~/DRKagi && git pull origin main
drkagi
git clone https://github.com/DRKagi/DRKagi.git
cd DRKagi
pip install -r requirements.txtdocker build -t drkagi .
docker run -it --env-file .env drkagi| Command | Description |
|---|---|
scan <IP> |
AI-powered smart scan |
autopilot <IP> |
Full 4-phase assessment |
autopilot <CIDR> |
Autopilot entire subnet |
| Any text | AI picks the right tool |
| Command | Description |
|---|---|
persona stealth |
👻 Ghost — maximum evasion |
persona aggressive |
⚡ Blitz — speed over stealth |
persona ctf |
🏴 CTF — flag hunting mode |
persona recon |
🔍 Recon — passive intelligence |
persona web |
🌐 Web — SQLi/XSS specialist |
persona off |
Reset to default |
| Command | Description |
|---|---|
write script <task> |
Generate Python script |
write script node <task> |
Generate Node.js script |
simulate <scenario> |
Model attack (no execution) |
wordlist <target> |
AI-generated targeted wordlist |
compliance <framework> |
PCI/HIPAA/ISO compliance check |
attack map |
Generate attack path diagram |
| Command | Description |
|---|---|
profile save <n> / load <n> |
Save/load engagement profiles |
session save <n> / load <n> |
Save/resume AI conversations |
vault add / list / export |
Credential vault |
show targets |
List discovered assets |
generate pdf |
Create PDF report |
dashboard |
Launch web UI |
| Command | Description |
|---|---|
status |
Show middleware + stats |
history |
Command history |
export md |
Export session to Markdown |
plugins / plugins reload |
Plugin management |
target <IP> |
Set active target (AI focuses on it) |
clear |
Clear terminal |
Switch between specialized AI personalities:
| Persona | Focus | Example |
|---|---|---|
| 👻 Ghost | Maximum stealth, -T0, decoys, proxychains | persona stealth |
| ⚡ Blitz | All ports, -T5, aggressive detection | persona aggressive |
| 🏴 CTF | Flags, default creds, GTFOBins | persona ctf |
| 🔍 Recon | Passive OSINT, no active scanning | persona recon |
| 🌐 Web | SQLi, XSS, SSRF, auth bypass | persona web |
console.print(f"[green]Sweeping {args}...[/green]")
Start the API server: python drkagi.py --api
| Endpoint | Method | Description |
|---|---|---|
/api/health |
GET | Health check |
/api/suggest |
POST | AI suggestion for a task |
/api/analyze |
POST | Analyze command output |
/api/script |
POST | Generate security script |
/api/targets |
GET | List discovered targets |
/api/simulate |
POST | Simulate attack scenario |
/api/cve |
GET | CVE lookup by service |
Contributions welcome! Open an issue or PR.
MIT License. See LICENSE for details.
Built with ❤️ for the security community
DRKagi v0.3.0 — Think like an attacker. Act like a professional.