RDKit MCP Server: Agentic Access to RDKit for LLMs
RDKit MCP Server is an open-source MCP server that enables language models to interact with RDKit through natural language. The goal is to provide agent-level access to every function in RDKit 2025.3.1 without writing any code.
Features
- Seamless Integration: Exposes RDKit functions via the Model Context Protocol (MCP).
- Language Model Support: Connect any LLM that supports the MCP protocol.
- CLI Client: Includes a command-line client powered by OpenAI for quick experimentation.
Table of Contents
Installation
Install the package:
pip install .
Usage
Start the Server
python run_server.py [--settings settings.yaml]
See settings.example.yaml for setting options
Once the server is running, any MCP-compliant LLM can connect. For example, see the Claude Desktop quickstart.
CLI Client
A CLI client is included for rapid prototyping with OpenAI:
export OPENAI_API_KEY="sk-proj-xxx"
python run_client.py
Available Tools
List all available RDKit tools exposed by the server:
python list_tools.py [--settings settings.yaml]
Evaluating with Promptfoo
We have provided examples configurations for using Promptfoo to evaluate the quality of RDKit tool outputs and agent responses against various models. The evals directory contains example configs and test cases.
Install Promptfoo
First, install Promptfoo globally using npm:
npm install -g promptfoo
Install RDKit mcp server
pip install .
Run Evaluation
To run an evaluation using the config in the evals directory:
cd evals
promptfoo eval
This will execute the tests defined in evals/promptfooconfig.yaml and report the results. You can customize the config and add your own test cases as needed.
The test results can be viewed using
promptfoo view
For more details, see the Promptfoo documentation.
Contributing
We welcome contributions, feature requests, and bug reports:
See CONTRIB.md for guidelines on how to get started.
Together, we can make RDKit accessible to a wider range of applications through natural language interfaces.