ninja1984/Chimera_C2

Chimera_C2 Framework A from-scratch C2 framework built to understand offensive security tooling internals. No copy-paste - every component written manually to learn how these systems actually work. Why I Built This Core learning goal: Build working implementations of C2 concepts with zero reliance on copy-pasted code or obvious signatures. If I can't explain every line, it doesn't go in. What's Actually Here This is a research project developed over 3 months of focused work. The framework currently implements: 98 Specialized Offensive Agents (rapidly expanding) Built from scratch with modular design. Each agent focuses on a specific red team task — from kernel diagnostics to advanced exfiltration. Tiered architecture - Separation between control logic and agent execution Modular components - Swappable communication methods and task handlers SQLite backend - Simple data persistence for testing/development The "AI integration" mentioned in earlier versions refers to experimental LLM-assisted script generation and analysis tools - not autonomous decision-making agents. Think Copilot-style assistance, not autonomous orchestration. Project Structure Chimera_C2/ ├── Agents/ # 98 agent implementations ├── Tier2_C_Agents/ # Secondary agent tier ├── Database/ # SQLite schemas and handlers ├── Modules/ # Task modules and communication handlers └── config/ # Environment configuration Current State This is active development work, not a polished product. The ~15 commits you see represent iterative rebuilding as I learned - many experiments got scrapped when I understood better approaches. What's committed works; what doesn't was deleted, not hidden. Technical Details Language: Python 3.x Database: SQLite (development), PostgreSQL planned Communication: HTTP/HTTPS with basic obfuscation Target platforms: Linux primarily, Windows agents in progress What I Learned How C2 frameworks manage agent registration and tasking Practical obfuscation techniques that actually evade basic detection Why most "AI-powered" security tools are just regex with marketing The gap between "works in the lab" and "works in production" Next Steps Staged payload delivery system Proper encryption for C2 traffic Windows agent hardening Detection evasion testing against real EDR ## Quick Start — How to Run Chimera C2 ### Recommended LLM Setup (Important) Chimera C2 was built and tested with **Ollama** using the following models: - `whiterabbitneo` - `obliteratus` These models give the best reasoning and red-team behavior for the multi-agent system. Make sure Ollama is running with one of these models before starting the orchestrator. ### 1. Environment Setup cd /path/to/Chimera_C2 # 1. Create and edit the environment variables cp config/.env.example config/.env # if the example file exists # OR create it manually: cat > config/.env << EOF C2_HOST=127.0.0.1 C2_PORT=9999 CALLBACK_HOST=192.168.56.101 # ← CHANGE to your host/VM IP CALLBACK_PORT=4444 GPU_IP=127.0.0.1 # ← your Ollama/GPU machine IP OLLAMA_HOST=127.0.0.1 EOF cd Orchestrator python3 live_orchestrator_backend.py cd Orchestrator python3 autonomous_orchestrator.py # Lab password used by demo agents (SQL Harvester, SSH Tunneler, Metasploit Bridge, etc.) LAB_PASSWORD=change_me_in_.env # ← Change this to your own secure password Disclaimer This is educational/research code for learning offensive security concepts. Not for unauthorized use. The goal is understanding defenses by building what they defend against.