Strapi Content

This MCP Server exposes your Strapi v5 content through the Model Context Protocol, enabling AI assistants to read, query, and manipulate content with safe, authenticated access. It provides a standard HTTP endpoint and a set of tools for CRUD operations and media handling, so you can integrate content workflows with conversational AI or automation clients.

How to use

You connect with your MCP client or AI assistant to the provided MCP endpoint or through a local subprocess that launches the MCP bridge. Once connected, you can list content types, query documents with filters and pagination, create or update documents, and upload media. Start by ensuring Strapi is running and the MCP server is reachable at the configured endpoint, then issue the available tool commands to read and write Strapi content.

How to install

Prerequisites: you need Node.js 18+ (including Node.js 22, 24+) and either npm or yarn installed.

Configuration and security

Install the MCP plugin for Strapi and enable it with a simple configuration. The plugin uses internal Strapi services with sanitization and can be adjusted for log level and security policies.

Troubleshooting

If you run into memory or connection issues, ensure you have enough memory allocated to the build, and verify that Strapi and the MCP endpoint are running on the expected port. Check the client configuration JSON for valid syntax and restart the client after changes.

Notes

The MCP endpoint is available at the HTTP URL shown in the configuration snippet. Use the local CLI or your AI client to interact with Strapi content through the defined tools.

Available tools

list_content_types

List all content types with their schemas.

list_components

List components with pagination.

find_many

Query documents with filters, sort, pagination.

find_one

Get single document by documentId.

create

Create a new document.

update

Update an existing document.

delete

Delete a document.

upload_media

Upload media from URL.