CStrike (Formally Ai-Driver)

An elite, modular offensive security automation framework with full TUI integration, OpenAI-enhanced command chaining, real-time recon, exploitation, and pivoting. Built for serious red team operations with support for split-VPN routing, ZAP/Burp Suite integration, Metasploit RPC automation, and interactive dashboards.

🚀 Key Features

Layered Reconnaissance with tools like nmap, dig, amass, subfinder, httpx, nikto, wafw00f, etc.
AI-Augmented Command Chaining using OpenAI (GPT-4o), streaming decisions in real time.
Auto-Triggered Exploit Chains: nuclei, ffuf, sqlmap, metasploit modules, smtp-user-enum, dnsenum.
Credential Loot Tracking & Reuse for brute-forcing (hydra, etc).
Metasploit RPC Automation via pymetasploit3 (auto-credentialed).
Burp/ZAP Integration with toggles via dashboard or AI.
TUI Dashboard (curses):
Live stream of current recon/exploit status
System metrics (CPU, RAM, VPN IP)
Service statuses
AI thought window
Split-screen log viewer with ERROR/WARN filters
Hotkeys to start/stop tools or view live logs
VPN Split Routing: Run tools as redteam user → traffic isolated via wg0 or tun0
Mission-Ready Modular Design: Easily add more tools or scan logic

📁 Project Structure

ai_driver/
├── ai_driver.py              # Main orchestrator
├── dashboard.py              # TUI live dashboard
├── setup_redteam_env.sh      # Environment + routing bootstrap
├── requirements.txt
├── .env                      # JSON config file (not dotenv)
├── modules/
│   ├── recon.py              # Multi-tool recon logic
│   ├── exploitation.py       # FFUF, nuclei, brute-force logic
│   ├── zap_burp.py           # ZAP/Burp scanner integration
│   ├── metasploit.py         # RPC control logic
│   ├── loot_tracker.py       # Tracks discovered usernames, creds, etc.
│   ├── ai_assistant.py       # OpenAI GPT assistant + command parser
│   └── utils.py              # Support utilities
├── results/                  # Output per target (loot, json, markdown)
└── logs/driver.log           # Global log (used by dashboard)

🧠 AI Features

Feature	Description
🧠 Thought Streaming	Dashboard shows real-time AI decisions as they are generated
⚡ Auto-Triggered Exploits	AI can auto-run chains like nuclei, hydra, metasploit modules
🤖 AI Post-Exploitation Loop	Follows up after initial exploit chain to suggest lateral moves
🧼 Safe Command Parser	Extracts only shell-safe commands from OpenAI replies
💾 Logs & Outputs Persisted	Stored in `results/<target>/ai_suggestions*.json` + logs

📡 Dashboard (TUI)

Hotkey	Function
`3`	Toggle Live Logs
`4`	Start Metasploit RPC, ZAP, Burp
`5`	Stop all services
`f`	Filter logs for `[ERROR]` / `[WARN]`
`q`	Quit the dashboard

Includes:

VPN IP detection via tun0 / wg0
CPU/RAM % usage
Current target
Phase progress (recon → AI → zap → metasploit → exploit)
AI “thought” log viewer
Log filter, scroll, highlight

🧪 Setup

1. Clone and configure

mkdir /opt/ai_driver
cd /opt
git clone https://github.com/culpur/cstrike.git
cd ai_driver
cp .env.example .env

2. Create virtualenv

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

⚙️ Split VPN Setup: redteam User

To isolate all scanning traffic through a VPN tunnel:

Step 1: Bootstrap with `setup_redteam_env.sh`

sudo bash setup_redteam_env.sh

Creates:

redteam user w/ /bin/zsh
Project files in /opt/ai_driver
ZSH aliases (ai_driver, setup_anon)
iptables + ip rule VPN routing (via wg0)
Default route in table wgvpn

Step 2: Test isolation

su - redteam
curl --interface wg0 https://ifconfig.me
ai_driver

All commands as redteam will now use the VPN route.

🧵 .env Example

{
  "target_scope": ["culpur.net"],
  "openai_api_key": "sk-xxxxxxxxxxxx",
  "allow_exploitation": true,
  "scan_modes": ["http", "dns", "port", "vulnscan"],
  "allowed_tools": [
    "nmap", "ffuf", "httpx", "sqlmap",
    "dig", "subfinder", "amass",
    "nikto", "wafw00f", "smtp-user-enum", "dnsenum"
  ],
  "max_threads": 10,
  "max_runtime": 300,
  "msf_username": "msf",
  "msf_password": "mypassword",
  "msf_host": "127.0.0.1",
  "msf_port": 55552,
  "zap_host": "127.0.0.1",
  "zap_port": 8090
}

🎯 Example Usage

# Launch full pipeline with live dashboard
python3 ai_driver.py

Will open curses dashboard
Shows status of recon/exploitation per target
Auto-invokes AI twice: post-recon + post-exploitation
Auto-triggers all exploit logic if allow_exploitation = true

🔮 Coming Soon

These features are already scaffolded or partially integrated:

✅ Proxy chaining logic for agent routing
✅ Pivot interface in TUI
✅ Credential heatmaps in dashboard
🔄 Remote agent registration
🔍 AI-driven lateral movement planner
📂 Export full report: Markdown + JSON

🔐 Legal

This tool is intended only for authorized red team use. Use against unauthorized targets is illegal and unethical.

📜 License

🙌 Credits & Contact

Crafted by Culpur Defense Inc.

GitHub: https://github.com/culpur
Website: https://culpur.net