MCP Server

The ElmapiCMS MCP Server connects AI-powered editors and agents to your CMS instance through the Model Context Protocol. Install it in Cursor, Claude Code, or any MCP-compatible tool and manage your collections, fields, content, and assets through natural language.

What It Does

The MCP server gives AI agents direct access to the ElmapiCMS API. Instead of manually creating collections and entering content, you can ask your AI editor to do it for you:

• Create and manage collections — Define collections with fields, validations, and field groups
• Manage fields — Add, update, and reorder fields within any collection
• Create and edit content — Create, update, list, and soft-delete content entries
• Manage assets — Upload, list, and delete media files
• Schema-aware — AI agents can read field type documentation and collection guidelines as built-in MCP resources, so they understand your content structure

Installation

Install the MCP server globally from npm:

npm install -g @elmapicms/mcp-server

Or install it locally in a project:

npm install @elmapicms/mcp-server

Configuration

The server requires three environment variables:

You can generate an API token from your project's Settings > API Access page. Make sure the token has the abilities needed for the tools you want to use (see Token Abilities below).

Setting Up in Cursor

Add the following to your Cursor MCP settings file (~/.cursor/mcp.json):

{
  "mcpServers": {
    "elmapicms": {
      "command": "npx",
      "args": ["@elmapicms/mcp-server"],
      "env": {
        "ELMAPI_API_URL": "https://your-domain.com/api",
        "ELMAPI_API_KEY": "your-api-key",
        "ELMAPI_PROJECT_ID": "your-project-uuid"
      }
    }
  }
}

After saving, go to Settings > MCP in Cursor and click the refresh button to start the server.

Setting Up in Claude Code

Add the MCP server using the Claude Code CLI:

claude mcp add elmapicms \
  -e ELMAPI_API_URL=https://your-domain.com/api \
  -e ELMAPI_API_KEY=your-api-key \
  -e ELMAPI_PROJECT_ID=your-project-uuid \
  -- npx @elmapicms/mcp-server

Available Tools

The MCP server exposes 17 tools across five categories:

Project

Collections

Fields

Content Entries

Assets

Advanced Queries

The list_entries tool supports powerful filtering through the where parameter. You can ask your AI editor to construct complex queries in natural language.

Supported operators: eq, lt, lte, gt, gte, not, like, in, not_in, null, not_null, between, not_between

Examples

Filter products under $50 that are in stock:

{
  "where": {
    "price": { "lt": 50 },
    "in-stock": true
  }
}

OR group — find entries tagged "clearance" or related to a campaign:

{
  "where": {
    "or": [
      { "tags": "clearance" },
      { "campaign": { "name": "Summer Sale" } }
    ]
  }
}

Sort by price, paginate 12 per page:

{
  "sort": "price:asc",
  "paginate": 12
}

Get just the count of matching entries:

{
  "where": { "status": "published" },
  "count": true
}

The server includes a built-in Query Reference resource with complete documentation on all operators, relation filtering, sorting, and pagination options.

Built-in Resources

The server exposes three reference resources that AI agents can read for context. These help agents understand your CMS without needing prior knowledge of ElmapiCMS:

• Field Types Reference (elmapicms://field-types) — Complete reference of all 16 field types, their options, validations, and common patterns.
• Collections Guide (elmapicms://collections-guide) — Guide for working with collections, singletons, reserved slugs, and best practices.
• Query Reference (elmapicms://query-reference) — Full documentation for content queries: where filters with 13 operators, OR groups, relation filtering, sorting, pagination, and examples.

Token Abilities

Your API token needs the appropriate abilities for the tools you want to use:

Using Multiple Projects

Each MCP server instance connects to one project. To work with multiple projects at the same time, add separate entries in your MCP config:

{
  "mcpServers": {
    "elmapicms-blog": {
      "command": "npx",
      "args": ["@elmapicms/mcp-server"],
      "env": {
        "ELMAPI_API_URL": "https://your-domain.com/api",
        "ELMAPI_API_KEY": "blog-project-api-key",
        "ELMAPI_PROJECT_ID": "blog-project-uuid"
      }
    },
    "elmapicms-store": {
      "command": "npx",
      "args": ["@elmapicms/mcp-server"],
      "env": {
        "ELMAPI_API_URL": "https://your-domain.com/api",
        "ELMAPI_API_KEY": "store-project-api-key",
        "ELMAPI_PROJECT_ID": "store-project-uuid"
      }
    }
  }
}

Example Workflow

Here is an example of what you can do with the MCP server in Cursor:

1. Ask Cursor: "Create a Blog Posts collection with title, slug, content, featured image, and published fields"
2. Cursor creates the collection and all fields with proper types, validations, and slug configuration
3. Ask Cursor: "Create a sample blog post about getting started with headless CMS"
4. Cursor creates the entry with generated content across all fields
5. Ask Cursor: "Build a Next.js page that fetches and displays blog posts from ElmapiCMS"
6. Cursor reads the collection schema and builds a fully typed frontend page

The AI agent handles the entire CMS setup while you focus on your frontend code.

Local Development

If your ElmapiCMS instance runs on a .test domain with a self-signed SSL certificate (e.g., via Laravel Herd), add NODE_TLS_REJECT_UNAUTHORIZED to your env config:

"env": {
  "ELMAPI_API_URL": "https://myproject.test/api",
  "ELMAPI_API_KEY": "your-api-key",
  "ELMAPI_PROJECT_ID": "your-project-uuid",
  "NODE_TLS_REJECT_UNAUTHORIZED": "0"
}

Links

• npm package
• GitHub repository
• Model Context Protocol