Drupal API MCP Server

A Model Context Protocol (MCP) server that provides semantic search across Drupal API documentation. Ask questions in natural language and get relevant Drupal functions, hooks, classes, and API documentation.

What is this?

MCP (Model Context Protocol) is a standard that allows AI assistants to access external tools and data sources. This server implements MCP to give AI assistants (like Cursor, Claude Desktop, etc.) access to Drupal's API documentation through semantic search.

The server automatically detects which Drupal version you're working with by reading your project's composer.lock file, ensuring you get documentation that matches your Drupal version.

Instead of manually searching docs, you can ask questions like:

• "How do I duplicate an entity?"
• "What hook fires when a node is saved?"
• "Show me the Entity API"

And get relevant Drupal API documentation with code examples for your specific Drupal version.

How It Works

This project uses vector embeddings to enable semantic search. Here's the architecture:

flowchart TB
    subgraph Maintainer["🔧 Maintainer Workflow"]
        direction LR
        A1[Drupal Core<br/>*.api.php files] -->|Parse| A2[API Documentation]
        A2 -->|Generate Embeddings| A3[Ollama<br/>Embedding Model]
        A3 -->|Export| A4[JSON Files<br/>in Git]
    end
    
    subgraph EndUser["👤 End User Workflow"]
        direction TB
        B1[JSON Files<br/>from Git] -->|Hydrate| B2[Redis<br/>Vector Store]
        B3[IDE<br/>Cursor/Claude] -->|Query| B4[MCP Server]
        B4 -->|Generate Query Embedding| B5[Ollama<br/>Embedding Model]
        B5 -->|Query Vector| B4
        B4 -->|Vector Search| B2
        B2 -->|Results| B4
        B4 -->|Documentation| B3
    end
    
    Maintainer -->|Pre-computed| EndUser
    
    style A1 fill:#16A085,stroke:#138D75,stroke-width:2px,color:#fff
    style A2 fill:#5C6BC0,stroke:#3F51B5,stroke-width:2px,color:#fff
    style A3 fill:#F39C12,stroke:#D68910,stroke-width:2px,color:#fff
    style A4 fill:#FF9800,stroke:#F57C00,stroke-width:2px,color:#fff
    style B1 fill:#FF9800,stroke:#F57C00,stroke-width:2px,color:#fff
    style B2 fill:#E74C3C,stroke:#C0392B,stroke-width:2px,color:#fff
    style B3 fill:#4A90E2,stroke:#2E5C8A,stroke-width:2px,color:#fff
    style B4 fill:#9B59B6,stroke:#6C3483,stroke-width:2px,color:#fff
    style B5 fill:#F39C12,stroke:#D68910,stroke-width:2px,color:#fff

The Process

1. Ingestion (Maintainers only):

   • Parse Drupal's *.api.php files to extract API documentation
   • Generate vector embeddings using Ollama's embedding model
   • Export embeddings to JSON files and commit to Git

2. Hydration (End Users):

   • Download pre-computed JSON embeddings from Git
   • Load them into Redis vector store for fast searching

3. Runtime (End Users):

   • Your IDE sends a natural language query to the MCP server
   • Server automatically detects your Drupal version from composer.lock (or uses a manually set version)
   • Server generates an embedding for your query using Ollama
   • Searches Redis vector store for similar documentation matching your Drupal version
   • Returns relevant API documentation with code examples for your specific version

Key Components

All components are open source and free:

• MCP PHP SDK (mcp/sdk): Handles MCP protocol communication between your IDE and the server
• PHP-Parser (nikic/php-parser): Parses Drupal's *.api.php files to extract functions, hooks, classes, and documentation
• DocBlock Extractor: Extracts documentation and code examples from PHP docblocks
• Docker & Docker Compose: Containerizes Redis and Ollama services for easy setup and management
• Redis (with Redis Stack): Vector database for fast semantic search of embeddings
• Ollama: Free, local LLM service for generating vector embeddings (no API keys required)
• Pre-computed Embeddings: JSON files containing Drup

... View full README on GitHub