DeepExploitFlow (DEF) is an automated penetration testing framework that combines traditional security tools with Reinforcement Learning (RL) and Large Language Models (LLMs).
Built on top of ExploitFlow, DEF aims to streamline vulnerability discovery, exploitation, and post-exploitation while reducing repetitive manual work.
- Automated reconnaissance and vulnerability scanning
- RL-based attack decision making
- AI-assisted payload and exploit generation
- Nuclei integration for web vulnerability scanning
- Support for vulnerabilities such as SQL Injection and OS Command Injection
- Modular and extensible architecture
- Performs IP, port, and service discovery
- Provides automated vulnerability scanning
- Supports customizable attack workflows
- Uses Reinforcement Learning to optimize testing decisions
- Integrates Nuclei for automated vulnerability scanning
- Uses a reward-based system to prioritize higher-impact findings
- Uses GPT-3.5 to assist with payload and exploit generation
- Applies prompt engineering for context-aware attack strategies
- Supports exploitation and post-exploitation workflows
git clone https://github.com/tronglinh23/DeepExploitFlow.git
cd DeepExploitFlowpip install -r requirements.txtexport OPENAI_API_KEY="<your-api-key>"Keep your API key private and never commit it to the repository.
DeepExploitFlow is intended for security research, education, and authorized penetration testing only.
Only use the framework against systems that you own or have explicit permission to test.
For questions, feedback, or contributions: