SitecoreMCP is a TypeScript-based MCP server that exposes a GraphQL API for interacting with Sitecore items, templates, and related data. It includes capabilities for item operations (get, create, update, delete), navigation (parent and ancestors), layout/site information, version control (item versions and statistics), and advanced features like schema scanning and natural language commands. You can leverage the included tools to query Sitecore data, perform mutations, and explore item dependencies. The server communicates via GraphQL and emits progress messages on stderr to support long-running operations in AI clients. To start using it, run the server locally or in your environment and query the GraphQL endpoint to perform actions such as fetching an item, listing children, or running a Sitecore scan.

Prerequisites:

• Node.js >= 18 installed on your machine
• Optional: Git and a build toolchain if you plan to build from source

Install and run locally:

1. Clone the repository: git clone https://github.com/GaryWenneker/SitecoreMCP.git cd SitecoreMCP

2. Install dependencies: npm install

3. Build the project (produces dist/index.js): npm run build

4. Run the MCP server: npm run start

   or directly

   node dist/index.js

5. Verify the GraphQL API is reachable (default http://localhost:3000/graphql).

Environment setup (optional):

• You can configure environment variables as needed (see additional_notes for common ones).
• If you customize the port, ensure it is allowed by your runtime environment.

Notes and tips:

• The server emits progress messages to stderr in the format [tool_name] Message... which is helpful for AI clients to display real-time status during long-running operations.
• Common environment variables to consider: PORT (server port), GRAPHQL_ENDPOINT (if you proxy), SITECORE_API_BASE (if you integrate with Sitecore APIs), and any authentication tokens your deployment requires.
• If you are building from source, ensure TypeScript is compiled to dist/index.js before running the server.
• When upgrading dependencies, check for breaking changes in the GraphQL schema and tool list (the server lists around 11–21 tools depending on version).
• For production deployments, consider containerization and setting appropriate resource limits and security practices for the GraphQL API.