MCP Server

Source: server/mcp_server.py

Overview

CatGo implements the Model Context Protocol (MCP), allowing external AI assistants (e.g., Claude Desktop) to interact with CatGo's tools for structure analysis, manipulation, and visualization.

Architecture

The MCP server wraps CatGo's Python server endpoints as MCP-compatible tools, exposing them through the standardized MCP protocol.

Exposed Tools

Structure Operations

• Parse structure files
• Query structure properties
• Generate slabs and supercells
• Optimize structures

Analysis

• Electronic structure (bands, DOS, COHP)
• MD analysis (RDF, RMSD, H-bonds)
• Symmetry detection

Workflow

• Create and run computational workflows

Setup

python server/main.py

Client Configuration

Add CatGo as an MCP server in your AI client's settings.

Point the client at the URL (no --mcp, no Python on the client):

http://localhost:8000/api/mcp/

Related

• MCP Server Tutorial
• REST API