Bacting MCP Server

A Model Context Protocol (MCP) server built with FastAPI for chemistry and cheminformatics support using pybacting.

Setup

1. Activate the virtual environment:

uv venv venv # only once
source venv/bin/activate

2. Install dependencies:

uv pip install -e .

Or with development dependencies:

uv pip install -e ".[dev]"

Running the Server

python server.py

The server will start on http://localhost:8000

Available Endpoints

• GET / - Server info
• GET /health - Health check
• POST /mcp/call - MCP tool calls
• GET /docs - Interactive API documentation (Swagger UI)

Example Usage

List available tools

curl -X POST http://localhost:8000/mcp/call \
 -H "Content-Type: application/json" \
 -d '{"method": "tools/list"}'

Call a tool

curl -X POST http://localhost:8000/mcp/call \
 -H "Content-Type: application/json" \
 -d '{"method": "tools/call", "params": {"name": "smiles_to_molecule", "arguments": {"smiles": "CCO"}}}'

Testing

To run the test suite:

uv pip install -e ".[dev]"
uv run pytest

Or run tests with verbose output:

uv run pytest tests/ -v

To run a specific test:

uv run pytest tests/test_server.py::test_smiles_to_molecule_valid -v

The test suite includes:

• Validation of available tools
• Testing valid SMILES string conversion (ethanol, benzene)
• Error handling for invalid SMILES strings
• Edge cases (empty strings, unknown tools)

Adding New Tools

To add new tools, modify the list_tools() and call_tool() functions in server.py.