An Open-Source Platform for Economic Simulation

• Declarative Shift: Define simulation complexity declaratively via YAML, separating the 'what' from the 'how' for readable and manageable simulations.
• AI-Native Brains: Integrate with Gemini, OpenAI, or local Ollama instances. Assign goals and personalities to agents for non-deterministic strategic decisions.
• Scientific Verifiability: Built with FastAPI and Pydantic, ensuring simulations are strictly determined by the scenario file and engine version for reproducibility.
• Case Study: Explore experimental models like the Hormuz Crisis, where AI agents used Propaganda as a Resource to mitigate political decay and achieve a ceasefire.
• Community-Driven Effort: Doxa is in active development, fostering a community for complex systems research.

Doxa is designed for modern researchers as a Python-native simulation environment. It allows for the simulation of nuanced human negotiation, irrational market sentiment, and adaptive trading strategies, even when agents are bound by rigid logic. The platform emphasizes scientific verifiability and reproducibility, making it a powerful tool for economic and complex systems research.