Overview
What is Strapi MCP Server?
A Model Context Protocol server for interacting with Strapi CMS. It enables AI assistants to interact with a Strapi instance through a standardized interface, supporting content types and REST API operations.
How to use Strapi MCP Server?
Install via npx in your Claude Desktop configuration. Create a configuration file at ~/.mcp/strapi-mcp-server.config.json with your Strapi instance's API URL and JWT token. Use the provided tools (e.g., strapi_rest, strapi_get_content_types, strapi_upload_media) to perform operations.
Key features of Strapi MCP Server
- Schema introspection
- REST API support with validation
- Media upload handling with format conversion
- JWT authentication
- Content type management
- Version compatibility management (v4 & v5)
- Write protection policy
Use cases of Strapi MCP Server
- Query and manage Strapi content types via an AI assistant
- Create, read, update, and delete content using the REST API
- Upload media with automatic format conversion and optimization
- Explore schema before performing operations to validate fields
FAQ from Strapi MCP Server
What Strapi versions are supported?
Both Strapi v4 and v5 are supported. Version can be specified in the config (e.g., "5.*", "4.1.5", "v4") and the server handles API differences automatically.
How do I authenticate with my Strapi instance?
Authentication is done using a JWT token. You must create an API token in your Strapi admin panel and add it to the configuration file.
Can I use this server with multiple Strapi instances?
Yes. You can configure multiple Strapi instances by adding multiple entries to the configuration file at ~/.mcp/strapi-mcp-server.config.json.
How does write protection work?
The server implements a strict write protection policy. All write operations (POST, PUT, DELETE, media upload) require explicit authorization. Each operation is logged and validated.
What are the main differences between Strapi v4 and v5 handled automatically?
v4 uses numeric IDs and nested attribute structures with a data wrapper, while v5 uses document-based IDs, flat data direct attribute access, and enhanced JWT security with integrated i18n.
