MiroFish
ActiveDescription
Multi-agent swarm intelligence engine that extracts seed information from the real world, constructs a high-fidelity parallel digital world with thousands of agents, and predicts future trajectories through social evolution simulations.
Key Features
- Multi-agent swarm simulation — thousands of agents with independent personalities and long-term memory freely interact
- GraphRAG knowledge graph — automatically constructs a high-fidelity parallel digital world from seed materials
- Dynamic memory — continuously updates individual and collective memory during simulation
- Dual-platform parallel — runs simulations on multiple platforms simultaneously for stable predictions
- ReportAgent toolset — deeply interacts with the simulated environment to generate detailed prediction reports
- Free interaction — chat with any agent in the simulated world to explore details
Use Cases
💡 Policy simulation and PR rehearsal — testing policies and PR strategies in zero-risk environments
💡 Financial market forecasting — predicting stock prices and public opinion trends from event signals
💡 Public opinion evolution analysis — simulating how viewpoints spread and collide on social media
💡 Creative narrative continuation — writing novel endings and exploring alternate fictional outcomes
💡 Strategic decision sandbox — enterprises running multi-round simulations before major events to optimize decisions
Strengths & Limitations
✅ Strengths
- • Actively maintained, recent updates
- • High community interest (71.9k stars)
- • Responsive to issues, low backlog
⚠️ Limitations
- • Restrictive license (AGPL-3.0)
Categories
Quick Start
git clone https://github.com/666ghj/MiroFish.git
cd MiroFish
cp .env.example .env # fill in LLM API keys
uv sync
uv run python main.py