Democratizing AI scientists with ToolUniverse
How to install tooluniverse?
mims-harvard/tooluniverse
$git clone https://github.com/mims-harvard/tooluniverseInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install tooluniverse by running `git clone https://github.com/mims-harvard/tooluniverse`, then use it for the current task and follow its documentation at https://github.com/mims-harvard/tooluniverse.
README.md
| 1 | # <img src="docs/_static/logo.png" alt="ToolUniverse Logo" height="28" style="vertical-align: middle; margin-right: 8px;" /> ToolUniverse: Democratizing AI scientists |
| 2 | |
| 3 | [](https://zitniklab.hms.harvard.edu/ToolUniverse/) |
| 4 | [](https://arxiv.org/abs/2509.23426) |
| 5 | [](https://badge.fury.io/py/tooluniverse) |
| 6 | [](https://registry.modelcontextprotocol.io) |
| 7 | [](https://aiscientist.tools) |
| 8 | [](https://join.slack.com/t/tooluniversehq/shared_invite/zt-3dic3eoio-5xxoJch7TLNibNQn5_AREQ) |
| 9 | [](https://aiscientist.tools/wechat) |
| 10 | [](https://www.linkedin.com/in/shanghua-gao-96b0b3168/) |
| 11 | [](https://x.com/ScientistTools) |
| 12 | [](https://pepy.tech/projects/tooluniverse) |
| 13 | |
| 14 | [//]: # (mcp-name: io.github.mims-harvard/tooluniverse) |
| 15 | |
| 16 | ## Install |
| 17 | |
| 18 | **AI agent (recommended)** — open your AI agent and run: |
| 19 | ``` |
| 20 | Read https://aiscientist.tools/setup.md and set up ToolUniverse for me. |
| 21 | ``` |
| 22 | The agent will walk you through MCP configuration, API keys, skill installation, and validation. |
| 23 | |
| 24 | <details> |
| 25 | <summary>or set up manually</summary> |
| 26 | |
| 27 | Add to your MCP config file: |
| 28 | ```json |
| 29 | { |
| 30 | "mcpServers": { |
| 31 | "tooluniverse": { |
| 32 | "command": "uvx", |
| 33 | "args": ["--refresh", "tooluniverse"], |
| 34 | "env": {"PYTHONIOENCODING": "utf-8"} |
| 35 | } |
| 36 | } |
| 37 | } |
| 38 | ``` |
| 39 | |
| 40 | Install agent skills: |
| 41 | ```bash |
| 42 | npx skills add mims-harvard/ToolUniverse |
| 43 | ``` |
| 44 | </details> |
| 45 | |
| 46 | **Python developers** — install the SDK: |
| 47 | ```bash |
| 48 | uv pip install tooluniverse |
| 49 | ``` |
| 50 | |
| 51 | **[`tu` CLI](https://zitniklab.hms.harvard.edu/ToolUniverse/guide/tu_cli.html)** — discover, inspect, run, and test tools from the terminal. |
| 52 | **[Python SDK](https://zitniklab.hms.harvard.edu/ToolUniverse/guide/python_guide.html)** — programmatic access for building AI scientist systems. |
| 53 | |
| 54 | ## Building AI Scientists with ToolUniverse |
| 55 | |
| 56 | <p align="center"> |
| 57 | <a href="https://www.youtube.com/watch?v=fManSJlSs60"> |
| 58 | <img src="https://github.com/user-attachments/assets/13ddb54c-4fcc-4507-8695-1c58e7bc1e68" width="600" /> |
| 59 | </a> |
| 60 | </p> |
| 61 | |
| 62 | *Click to watch the demo* [(YouTube)](https://www.youtube.com/watch?v=fManSJlSs60) [(Bilibili)](https://www.bilibili.com/video/BV1GynhzjEos/?share_source=copy_web&vd_source=b398f13447281e748f5c41057a2c6858) |
| 63 | |
| 64 | ## What is ToolUniverse? |
| 65 | |
| 66 | ToolUniverse is an ecosystem for creating AI scientist systems from any large language model. Powered by the AI-Tool Interaction Protocol, it standardizes how LLMs identify and call tools, integrating more than **1000 machine learning models, datasets, APIs, and scientific packages** for data analysis, knowledge retrieval, and experimental design. |
| 67 | |
| 68 | **Key features:** |
| 69 | |
| 70 | - [**AI-Tool Interaction Protocol**](https://zitniklab.hms.harvard.edu/ToolUniverse/guide/interaction_protocol.html): Standardized interface governing how AI scientists issue tool requests and receive results |
| 71 | - [**Universal AI Model Support**](https://zitniklab.hms.harvard.edu/ToolUniverse/guide/building_ai_scientists/index.html): Works with Claude, GPT, Gemini, Qwen, Deepseek, and open models |
| 72 | - [**MCP Integration**](https://zitniklab.hms.harvard.edu/ToolUniverse/guide/building_ai_scientists/mcp_support.html): Native Model Context Protocol server with configurable transport and tool selection |
| 73 | - [**Async Operations**](https://zitniklab.hms.harvard.edu/ToolUniverse/expand_tooluniverse/async_tools_guide.html): Long-running tasks (protein docking |
Related
6 picksType
- wikipedia-mcpMCP server for Wikipedia APIMay 2026
- gpt-researcherGPT Researcher is an autonomous agent designed for comprehensive online research on a variety of tasks.Jul 2026
- arxiv-mcp-serverMCP server for accesing arXiv APIAug 2025
- arxiv-mcp-serverA Model Context Protocol server for searching and analyzing arXiv papersJul 2026
- knowledgebase-mcpBioContextAI Knowledgebase MCP server for biomedical agentic AIJan 2026
- mcp-server-serperSerper MCP Server supporting search and webpage scrapingMar 2025
