A multi-headed digital sentinel for OSINT, security, and safety.
Project Cerberus is a modular security and intelligence platform designed to operate locally/off-grid with a focus on OSINT, security monitoring, and safety tools. The system is built with a "multi-headed" architecture, where each "head" specializes in different aspects of security and intelligence gathering.
- Modular Architecture: Multiple specialized components that work together or independently
- Local-First Operation: Designed to run entirely on your own hardware
- Off-Grid Capability: Can operate without internet connectivity
- LLM Integration: Leverages local LLMs for intelligence analysis
- Visual Analysis: Process and analyze images and visual data
- OSINT Capabilities: Gather and analyze open-source intelligence
- Customizable Dashboard: Monitor and control all components from a central interface
- AI Agents: Specialized agents for various security tasks
The Cerberus project consists of the following main components:
Core infrastructure and LLM integration with Docker stack for Ollama, FastAPI gateway, and OpenWebUI.
OSINT and reconnaissance capabilities with modules for username, breach, and domain intelligence gathering.
The main repository structure and Ocean-style dashboard for monitoring and controlling the Cerberus system.
Collection of specialized AI agents for various security tasks:
- Recon Intake Agent
- Chroma Visual Tools Agent
- Call Sentinel Triage Agent
- Deal Tracker Agent
- OSINT Report Summarizer Agent
- Docker and Docker Compose
- Python 3.8+
- Node.js (for dashboard)
- Git
- 8GB+ RAM recommended (16GB+ for larger models)
- 20GB+ free disk space
-
Clone the component repositories
git clone https://github.com/Stormchaser407/cerberus-aegis.git git clone https://github.com/Stormchaser407/cerberus-recon.git git clone https://github.com/Stormchaser407/cerberus-watch.git git clone https://github.com/Stormchaser407/cerberus-agents.git
-
Set up the core Aegis infrastructure
cd cerberus-aegis # Follow the setup instructions in the Aegis README
-
Configure the Recon Engine
cd cerberus-recon # Follow the setup instructions in the Recon README
-
Deploy the Watch Dashboard
cd cerberus-watch # Follow the setup instructions in the Watch README
-
Import the Agents into MindStudio
cd cerberus-agents # Follow the setup instructions in the Agents README
The Cerberus system is designed with a modular architecture:
- Aegis: Core infrastructure and LLM integration
- Recon: OSINT and reconnaissance capabilities
- Wraith: Stealth monitoring and tracking
- Oren: Threat intelligence and analysis
- Mavrakis: Social engineering and persona management
- Chroma: Visual analysis and media processing
- Local-First: Designed to run entirely on your own hardware
- Off-Grid Capable: Can operate without internet connectivity
- Modular Design: Use only the components you need
- Themes: Choose between "Hacker Mode" and "Nosy Neighbor Mode"
- Data Sovereignty: All data remains under your control
- Privacy-Focused: No external API dependencies required
- Backend: FastAPI
- Frontend: React
- LLM Integration: Ollama (LLaMA 3), GPT4All, LangChain
- Hosting: Docker, docker-compose, multi-stage builds
- Mobile Support: Termux on Android
Contributions are welcome! Here's how you can contribute:
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add some amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Please ensure your code follows the project's coding standards and includes appropriate tests.
This project is licensed under the MIT License - see the LICENSE file for details.
- Stormchaser407 - Initial work - GitHub
- The open-source community for providing tools and libraries
- Contributors to the LLM ecosystem
- Security researchers and OSINT practitioners