What is ChemMCP?
ChemMCP is an easy-to-use and extensible chemistry toolkit for LLMs and AI assistants, compatible with the Model Context Protocol (MCP). By integrating powerful chemistry tools, ChemMCP can make general AI models capable of chemistry capabilities, performing molecular analysis, property prediction, and reaction synthesis tasks, without requiring domain-specific training. ChemMCP can also be easily integrated in your research for data processing, agentic applications, and more. The ChemMCP toolset is largely extended and updated from our prior paper ChemToolAgent, and will be continuously updated and expanded in the future.
Join our Discord community to discuss ChemMCP, get help, and build together!
Key features
🔌 Plug-and-Play Chemistry Tools for AI Assistants: ChemMCP tools can be integrated into any MCP-enabled LLM clients in just minutes, allowing researchers to augment LLMs with chemistry capabilities without additional training.
🛠️ Standalone Toolkit for Custom Workflows: With its decoupled design and unified interfaces, ChemMCP tools can be easily imported into your workflow, to process data, assemble pipeline steps, or build bespoke agentic applications — via MCP or Python, whichever you prefer.
🤖 Native RL Agent Framework: ChemMCP natively supports multi-turn interactive loops between agents and tool, providing an ideal infrastructure and testbed for scientific tool-using agents.
📦 Modular and Extensible Design: Adding a new tool is as simple as writing a Python file (see here). All tools follow a consistent schema, ensuring clear interfaces, easy maintenance, and automatic documentation.
Where to Start?
Follow the steps below to embrace ChemMCP:
2. Explore Tools
: Browse our extensive collection of chemistry tools and their detailed documentation.
3. Configure with QuickConfig : Easily customize your tool selection and configuration with our easy-to-use utility.
4. Contribute to ChemMCP
: Join our community by implementing new tools following our development guide, or help improve ChemMCP by reporting issues and suggesting features on GitHub
.
About Us
ChemMCP is an open-source project developed by the OSU NLP Group and led by Botao Yu. We're pioneering the integration of advanced chemistry tools with large language models, bridging the gap between computational chemistry and AI. Our vision is to democratize access to sophisticated chemistry tools, empowering researchers and practitioners to tackle complex scientific challenges with the assistance of AI technology.
Citations
If ChemMCP is valuable to your research or development, please kindly cite our works:
@misc{yu2025chemmcp,
author = {Botao Yu and Huan Sun},
title = {ChemMCP: An Easy-to-Use and Extensible Toolkit for Chemistry Agents},
year = {2025},
url = {https://osu-nlp-group.github.io/ChemMCP/},
note = {2025-06-05-01}
}
@article{yu2024chemtoolagent,
title={ChemToolAgent: The Impact of Tools on Language Agents for Chemistry Problem Solving},
author={Botao Yu and Frazier N. Baker and Ziru Chen and Garrett Herb and Boyu Gou and Daniel Adu-Ampratwum and Xia Ning and Huan Sun},
journal={arXiv preprint arXiv:2411.07228},
year={2024}
}Contact
Have questions or feedback?
•
Join our Discord community to discuss ChemMCP, get help, and connect with other users!
•
Open an issue for bug reports or feature requests on our GitHub repository.
•
Email us at yu.3737@osu.edu – we are eager to know your ideas and suggestions!