Contentful Delivery MCP Server

A Model Context Protocol (MCP) server that provides seamless access to Contentful's Delivery API through AI assistants. Query and retrieve content entries, assets, and content types using natural language.

Quick Start

Install the package in your project:

npm install @mshaaban0/contentful-delivery-mcp-server

Or globally:

npm install -g @mshaaban0/contentful-delivery-mcp-server

Set up your Contentful credentials:

export CONTENTFUL_SPACE_ID="your_space_id"
export CONTENTFUL_ACCESS_TOKEN="your_access_token"
# Optional: Restrict content to specific content types
export CONTENTFUL_CONTENT_TYPE_IDS="blogPost,article,product"

Features

Natural language queries to search content
Retrieve entries by ID or content type
Asset management
Content type schema access
Pagination support
Rich text content handling

Available Tools

query_entries - Natural language search across all content
get_entry - Fetch specific entry by ID
get_entries - List entries with filtering
get_assets - Browse all assets
get_asset - Get asset details by ID
get_content_type - View content type schema
get_content_types - List available content types

Integration with Mastra AI

Mastra AI provides seamless integration with this MCP server. Here's how to set it up:

import { MastraMCPClient } from "@mastra/mcp";
import { Agent } from "@mastra/core/agent";

// Initialize the MCP client
const contentfulClient = new MastraMCPClient({
  name: "contentful-delivery",
  server: {
    command: "npx",
    args: ["-y", "@mshaaban0/contentful-delivery-mcp-server@latest"],
    env: {
      CONTENTFUL_ACCESS_TOKEN: "your_access_token",
      CONTENTFUL_SPACE_ID: "your_space_id",
      // Optional: Restrict content to specific content types
      CONTENTFUL_CONTENT_TYPE_IDS: "blogPost,article,product"
    }
  }
});

// Create an AI agent with access to Contentful
const assistant = new Agent({
  name: "Content Assistant",
  instructions: `
    You are a helpful assistant with access to our content database.
    Use the available tools to find and provide accurate information.
  `,
  model: "gpt-4",
});

// Connect and register tools
await contentfulClient.connect();
const tools = await contentfulClient.tools();
assistant.__setTools(tools);

// Example usage
const response = await assistant.chat("Find articles about machine learning");

Development

# Clone the repo
git clone https://github.com/mshaaban0/contentful-delivery-mcp-server.git

# Install dependencies
npm install

# Build
npm run build

# Development with auto-rebuild
npm run watch

# Run the inspector
npm run inspector

Debugging

The MCP Inspector provides a web interface for debugging:

npm run inspector

Visit the provided URL to access the debugging tools.

Resources

License

MIT