MiroFish MCP

简洁通用的群体智能引擎，预测万物 A Simple and Universal Swarm Intelligence Engine, Predicting Anything

English | 中文文档

🔌 This is a fork: MiroFish as an agent-native MCP server

This repository is a fork of 666ghj/MiroFish. The simulation engine, backend, frontend and research behind it are the work of the original MiroFish team — this fork does not change any of it.

What this fork adds is an integration layer that turns MiroFish from a web application you operate by hand into an engine your AI assistant can drive directly, running entirely on your own machine.

What it unlocks

Concretely, you can now say "simulate how students react to this tuition policy, then tell me the three most likely outcomes" and have an agent carry out the whole pipeline.

How to use it

Give START_HERE.md to your AI agent. It contains a single prompt to paste; the agent sets everything up and tells you what it needs from you.

Then read user_guide.md — what to ask for, how to interview agents mid-run, how to control cost, and the process-lifetime behaviour that catches people out. docs/AGENT_SETUP.md has the per-client configuration details.

What was added

The nine MCP tools cover project and ontology creation, graph building, task polling, persona synthesis and simulation launch, live status, mid-run agent interviews, stopping a run, report generation, and follow-up Q&A against the simulation graph.

Notes

• Requires Python 3.11 or 3.12 (backend/pyproject.toml pins >=3.11,<3.13). The bootstrap fetches a suitable interpreter via uv if you do not have one — no admin rights needed.

• You supply your own API keys. An OpenAI-compatible LLM endpoint and a Zep key, both in .env. Nothing is bundled.

• Simulations cost money. Every agent makes LLM calls every round, so cost scales with personas × rounds. Start small.

• Licence: MiroFish is AGPL-3.0 and this fork remains so. If you run a modified version as a network service, you must offer its source.

Everything below this point is the original project's documentation.

Related MCP server: ASTRA MCP Server

⚡ Overview

MiroFish is a next-generation AI prediction engine powered by multi-agent technology. By extracting seed information from the real world (such as breaking news, policy drafts, or financial signals), it automatically constructs a high-fidelity parallel digital world. Within this space, thousands of intelligent agents with independent personalities, long-term memory, and behavioral logic freely interact and undergo social evolution. You can inject variables dynamically from a "God's-eye view" to precisely deduce future trajectories — rehearse the future in a digital sandbox, and win decisions after countless simulations.

You only need to: Upload seed materials (data analysis reports or interesting novel stories) and describe your prediction requirements in natural language MiroFish will return: A detailed prediction report and a deeply interactive high-fidelity digital world

Our Vision

MiroFish is dedicated to creating a swarm intelligence mirror that maps reality. By capturing the collective emergence triggered by individual interactions, we break through the limitations of traditional prediction:

• At the Macro Level: We are a rehearsal laboratory for decision-makers, allowing policies and public relations to be tested at zero risk

• At the Micro Level: We are a creative sandbox for individual users — whether deducing novel endings or exploring imaginative scenarios, everything can be fun, playful, and accessible

From serious predictions to playful simulations, we let every "what if" see its outcome, making it possible to predict anything.

🌐 Live Demo

Welcome to visit our online demo environment and experience a prediction simulation on trending public opinion events we've prepared for you: mirofish-live-demo

📸 Screenshots

🎬 Demo Videos

1. Wuhan University Public Opinion Simulation + MiroFish Project Introduction

Click the image to watch the complete demo video for prediction using BettaFish-generated "Wuhan University Public Opinion Report"

2. Dream of the Red Chamber Lost Ending Simulation

Click the image to watch MiroFish's deep prediction of the lost ending based on hundreds of thousands of words from the first 80 chapters of "Dream of the Red Chamber"

Financial Prediction, Political News Prediction and more examples coming soon...

🔄 Workflow

1. Graph Building: Seed extraction & Individual/collective memory injection & GraphRAG construction

2. Environment Setup: Entity relationship extraction & Persona generation & Agent configuration injection

3. Simulation: Dual-platform parallel simulation & Auto-parse prediction requirements & Dynamic temporal memory updates

4. Report Generation: ReportAgent with rich toolset for deep interaction with post-simulation environment

5. Deep Interaction: Chat with any agent in the simulated world & Interact with ReportAgent

🚀 Quick Start

Option 1: Source Code Deployment (Recommended)

Prerequisites

1. Configure Environment Variables

# Copy the example configuration file
cp .env.example .env

# Edit the .env file and fill in the required API keys

Required Environment Variables:

# LLM API Configuration (supports any LLM API with OpenAI SDK format)
# Recommended: Alibaba Qwen-plus model via Bailian Platform: https://bailian.console.aliyun.com/
# High consumption, try simulations with fewer than 40 rounds first
LLM_API_KEY=your_api_key
LLM_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
LLM_MODEL_NAME=qwen-plus

# Zep Cloud Configuration
# Free monthly quota is sufficient for simple usage: https://app.getzep.com/
ZEP_API_KEY=your_zep_api_key

2. Install Dependencies

# One-click installation of all dependencies (root + frontend + backend)
npm run setup:all

Or install step by step:

# Install Node dependencies (root + frontend)
npm run setup

# Install Python dependencies (backend, auto-creates virtual environment)
npm run setup:backend

3. Start Services

# Start both frontend and backend (run from project root)
npm run dev

Service URLs:

• Frontend: http://localhost:3000

• Backend API: http://localhost:5001

Start Individually:

npm run backend   # Start backend only
npm run frontend  # Start frontend only

Option 2: Docker Deployment

# 1. Configure environment variables (same as source deployment)
cp .env.example .env

# 2. Pull image and start
docker compose up -d

Reads .env from root directory by default, maps ports 3000 (frontend) / 5001 (backend)

Mirror address for faster pulling is provided as comments in docker-compose.yml, replace if needed.

📬 Join the Conversation

 

The MiroFish team is recruiting full-time/internship positions. If you're interested in multi-agent simulation and LLM applications, feel free to send your resume to: mirofish@shanda.com

📄 Acknowledgments

MiroFish has received strategic support and incubation from Shanda Group!

MiroFish's simulation engine is powered by OASIS (Open Agent Social Interaction Simulations), We sincerely thank the CAMEL-AI team for their open-source contributions!

📈 Project Statistics