home / mcp / chemmcp mcp server

ChemMCP MCP Server

Provides a Python-based MCP server hosting 19 chemical tools for computation, analysis, and AI-assisted generation.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "cassiusoat-chemmcp_lite": {
      "url": "https://mcp.chem.example.com/mcp",
      "headers": {
        "LLM_MODEL_NAME": "openai/gpt-4o",
        "OPENAI_API_KEY": "your_openai_api_key_here",
        "TAVILY_API_KEY": "your_tavily_api_key_here",
        "CHEMSPACE_API_KEY": "your_chemspace_api_key_here"
      }
    }
  }
}

ChemMCP is a Python-based MCP server that bundles 19 specialized chemical molecule tools into a single, task-oriented platform. It enables you to compute properties, analyze structures, and generate descriptions or new molecules using AI, all accessible through a unified MCP interface.

How to use

You run the MCP server locally and connect your MCP client to it to access all 19 chemical tools. Start the server with the complete command from the configuration snippet, then specify the tools you want to invoke in your client requests. The server handles tasks such as molecule weight calculations, SMILES validation, IUPAC conversions, and AI-assisted molecule descriptions or generation.

How to install

Prerequisites you need before installation:

conda create -n chemmcp python=3.11
conda activate chemmcp

Install core dependencies and the package in editable mode to enable local development and testing.

pip install -r requirements.txt
```
```
pip install -e .

Choose an installation option that matches your needs.

# Option 1: Basic installation (recommended, core deps, 17 tools)
pip install -r requirements.txt
```
```
# Option 2: Full installation (including AI tools)
# A: Optional dependencies group
pip install -e ".[ai]"

# B: Manual AI dependencies
pip install transformers>=4.20.0 torch>=1.9.0

# Option 3: Development environment
pip install -e ".[all,dev]"

Environment and prerequisites

Set up environment variables for tools that require external APIs. These keys are used by network search, LLM-based tooling, and optional databases.

export TAVILY_API_KEY="your_tavily_api_key"
export LLM_MODEL_NAME="openai/gpt-4o"
export OPENAI_API_KEY="your_openai_api_key"
export CHEMSPACE_API_KEY="your_chemspace_api_key"  # optional

Available tools

WebSearch

Executes web searches to retrieve chemical information and relates links.

MoleculeWeight

Calculates molecular weight from a SMILES structure.

MoleculeAtomCount

Counts the number of each atom type in a SMILES structure.

MoleculeSimilarity

Computes Tanimoto similarity between two SMILES strings.

FunctionalGroups

Identifies functional groups present in a molecule.

SmilesCanonicalization

Converts SMILES to a canonical standardized form.

Iupac2Smiles

Converts IUPAC names to SMILES strings.

Smiles2Iupac

Converts SMILES strings to IUPAC names.

Smiles2Formula

Converts SMILES to molecular formula.

Name2Smiles

Converts common chemical names to SMILES.

Selfies2Smiles

Converts SELFIES to SMILES.

Smiles2Selfies

Converts SMILES to SELFIES.

Smiles2Cas

Converts SMILES to CAS registry numbers.

MoleculeSmilesCheck

Validates SMILES syntax.

ReactionSmilesCheck

Validates reaction SMILES syntax.

MoleculeVisualizer

Generates 2D molecular structure diagrams.

MoleculeCaptioner

Generates AI-based textual descriptions of molecules.

MoleculeGenerator

Generates molecular structures from textual descriptions using AI.

MoleculeModifier

Mods molecules to create derivatives and modifications.