strapi-plugin
Strapi plugin that integrates Model Context Protocol (MCP) functionality, enabling AI models to interact with your Strapi content and system information through a standardized protocol
claude mcp add --transport stdio virtuslab-open-source-strapi-plugin-mcp npx -y @sensinum/strapi-plugin-mcp
How to use
This MCP server is a Strapi plugin that exposes Model Context Protocol capabilities for AI models to interact with Strapi content and system information. It provides a streamable HTTP transport at the endpoint /api/mcp/streamable, which enables MCP clients to initialize, send requests, and terminate sessions in a session-aware manner. Through this plugin, AI assistants can explore content types, retrieve system information, call Strapi services, and manage sessions either in memory or via Redis. The available MCP tools are delivered over a streamable HTTP channel, making it compatible with clients like Claude Desktop, Cursor, and other MCP ecosystems. You can configure the MCP client for Claude Desktop or Cursor by pointing it to the streamable endpoint and, where applicable, supplying a note for how to connect. The plugin supports both memory-based sessions for development and Redis-backed sessions for production deployments, with optional configuration of Redis connection details and session TTLs.
How to install
Prerequisites:
- Strapi v5.0.0 or higher
- Node.js v18.0.0 or higher (recommended: v20 LTS)
- npm, yarn, or pnpm as package manager
- Optional Redis server if you plan to use Redis-backed sessions
Installation steps:
- Install Strapi (if not already installed) and create your project as usual.
- Install the MCP plugin:
- npm: npm install @sensinum/strapi-plugin-mcp
- yarn: yarn add @sensinum/strapi-plugin-mcp
- pnpm: pnpm add @sensinum/strapi-plugin-mcp
- Enable and configure the plugin inside Strapi’s config:
- Create or edit config/plugins.js (or config/plugins.ts) and add the MCP configuration under mcp with your preferred session type (memory or redis).
- Start Strapi. The MCP endpoints will be available at:
- GET /api/mcp/streamable
- POST /api/mcp/streamable
- DELETE /api/mcp/streamable
- If you plan to use Redis-based sessions, ensure Redis is running and configure the connection details in config/plugins.js as described in the plugin’s documentation.
Note: After installation, you may need to restart your Strapi server for the plugin to initialize properly.
Additional notes
- Security: This plugin exposes internal Strapi functionality and should NEVER be enabled in production environments. It is intended for development and local use only.
- Endpoints: The MCP streamable endpoints support session-based communication with automatic session management.
- Session options: You can choose memory-based sessions for development or Redis-based sessions for production. Configure ttlMs, maxSessions (memory), and Redis connection options as described in the plugin documentation.
- IP allowlisting: If needed, restrict access to MCP endpoints via an allowedIPs configuration in your Strapi plugins config.
- Compatibility: Requires Strapi v5.0.0+ and Node.js v18+; ensure your environment matches these prerequisites before installing.
- Restart: After enabling or changing plugin configuration, restart Strapi to apply changes.
Related MCP Servers
mcp-graphql
Model Context Protocol server for GraphQL
leetcode
An MCP server enabling automated access to LeetCode's problems, solutions, and public data with optional authentication for user-specific features, supporting leetcode.com & leetcode.cn sites.
rohlik
MCP server that lets you shop groceries across the Rohlik Group platforms (Rohlik.cz, Knuspr.de, Gurkerl.at, Kifli.hu, Sezamo.ro)
hackmd
A Model Context Protocol server for integrating HackMD's note-taking platform with AI assistants.
mcp-konnect
A Model Context Protocol (MCP) server for interacting with Kong Konnect APIs, allowing AI assistants to query and analyze Kong Gateway configurations, traffic, and analytics.
google-scholar
An MCP server for Google Scholar written in TypeScript with Streamable HTTP