AgentFarm

Simulation and analysis platform for agent-based modeling, reinforcement learning experiments, and complex adaptive systems research in the Dooders project.

Note: Active development — APIs may change between releases. See CHANGELOG.

Features

• Multi-agent simulations with configurable genomes, actions, and spatial indexing
• Experiment runner, SQLite-backed metrics, and analysis pipeline
• RL/decision stack (DQN, distillation, evolution experiments)
• REST/WebSocket API and structured logging with structlog

Quick start

git clone https://github.com/Dooders/AgentFarm.git && cd AgentFarm
python -m venv venv && source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt && pip install -e .
python run_simulation.py --environment development --steps 1000

Results appear under simulations/. See Installation and First simulation for more.

Documentation

• Docs site: dooders.github.io/AgentFarm
• Hub: docs/README.md
• Research: Devlog · Experiments
• Contributing: CONTRIBUTING.md

License

Apache 2.0 — see LICENSE.

Support

Open an issue on GitHub.