Agent skill · research science · aradotso
mirofish-swarm-intelligence
mirofish-swarm-intelligence — an agent skill for coding agents.
Why this skill is useful
Provides specific commands and configurations for setting up and running the MiroFish swarm intelligence engine, which are not widely known.
What it needs
Requires node.js, python, uv installed locally. Requires zep account access. About 7k tokens when loaded. Last updated 2026-07-09. 70 stars on the source repository.
What this skill does
Seed Material (text/data) ↓ [Graph Construction] — Entity extraction, GraphRAG, memory injection ↓ [Environment Setup] — Agent persona generation, relationship mapping ↓ [Dual-Platform Simulation] — Parallel agent interaction + temporal memory updates ↓ [Report Generation] — ReportAgent with toolset queries simulation world ↓ [Deep Interaction] — Chat with any agent or ReportAgent bash curl -LsSf https://astral.sh/uv/install.sh sh bash git clone https://github.com/666ghj/MiroFish.git cd MiroFish Copy environment template cp .env.example .env env LLM — must be OpenAI SDK compatible Recommended: Alibaba Qwen via DashScope LLMAPIKEY=$LLMAPIKEY LLMBASEURL=https://dashscope.aliyuncs.com/compatible-mode/v1 LLMMODELNAME=qwen-plus Zep Cloud — agent long-term memory Free tier at https://app.getzep.com/ ZEPAPIKEY=$ZEPAPIKEY env OpenAI LLMBASEURL=https://api.openai.com/v1 LLMMODELNAME=gpt-4o Local Ollama LLMBASEURL=http://localhost:11434/v1 LLMMODELNAME=llama3.1 LLMAPIKEY=ollama bash One-command full install (Node + Python) npm run setup:all bash npm run setup # Node.js deps (root + frontend) npm run setup:backend # Python deps via uv (creates .venv automatically) bash From project root — starts frontend (3000) and backend (5001) npm run dev bash npm run backend # Flask API on http://localhost:5001 npm run frontend # React UI on http://localhost:3000 bash cp .env.example .env Edit .env with your keys docker compose up -d Frontend: http://localhost:3000 Backend: http://localhost:5001 python Backend uses ZEPAPIKEY from environment Each agent gets a unique Zep session for long-term memory Memory is updated each round with new interactions python import requests response = requests.post("http://localhost:5001/api/simulation/start", json={ "seedmaterial": "Your seed text here — news article, report, or story excerpt", "predictionquery": "What will public opinion look like in 30 days?", "numagents": 100, "maxrounds": 30 }) simulationid = response.json()["simulationid"] print(f"Simul …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aradotso/mirofish-swarm-intelligence