GitHub Repository

A YAML-driven multi-agent simulation platform for economic and social systems. It combines LLM-backed agents, market microstructure, relation graphs, and world events behind a FastAPI API and a React client.

12 starsPython

Doxa – Open-source emergent simulator for geopolitical scenarios

Name: Doxa – Open-source emergent simulator for geopolitical scenarios
Availability: InStock
Author: dinarino

by dinarino·Apr 24, 2026·2 points·0 comments

Visit Project View on HN

AI Analysis

●MidBig BrainShip It

YAML-driven LLM agent simulation competing directly with AutoGen and AgentScope.

Strengths

•Whitepaper with DOI and Colab notebook shows genuine research intent, not just code dump.
•Pre-built scenarios cover Hormuz crisis, financial markets, and AI treaty negotiation out of box.

Weaknesses

•Multi-agent simulation space is crowded; AgentScope, AutoGen, and LangGraph already dominate.
•Only 4 GitHub stars and 2 issues suggest minimal real-world testing or adoption so far.

Post Description

Hi! We, Vincenzo and Riccardo, built Doxa as an agnostic engine for emergent simulations with agents for constrainted scenarios (like geopolitical, economics, ...) and work well with LLMs like Qwen2.5:7B, Llama but also cloud models such as Gemini.

Doxa can be configured in minutes with YAML scenarios, where you can assign agents' persona, portfolios, constraints.

Here a ready-to-use Colab: https://colab.research.google.com/github/VincenzoManto/Doxa/...

Let us know your thoughts!