MCP GraphQL Server

Provides dynamic access to GraphQL endpoints via MCP with schema introspection and query execution.
  • typescript

0

GitHub Stars

typescript

Language

4 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

You can use the mcp-graphql MCP server to let large language models interact with GraphQL APIs. It can introspect a GraphQL schema and execute queries, enabling dynamic discovery and use of GraphQL endpoints from within your MCP-enabled workflow.

How to use

You run the mcp-graphql server and connect it through an MCP client to enable your models to query GraphQL endpoints. The server can discover the GraphQL schema automatically via introspection or consume a local schema file. By default, mutations are disabled for safety, so restrict edits to your data unless you explicitly enable mutations.

How to install

Prerequisites: ensure you have Node.js and npm installed on your system.

Install and run the server via an inline command, or configure it through a JSON MCP setup as shown in the examples.

Install and run using the command line directly to start a local GraphQL MCP instance. You can point it at your GraphQL endpoint and optionally provide headers or a local schema.

Additional notes

Configuration options include specifying the GraphQL endpoint, optional HTTP headers for requests, whether to enable mutations, and whether to use a local schema file instead of performing an introspection. Mutations are disabled by default to protect your data.

Available tools

introspect-schema

Retrieves the GraphQL schema by using either a local schema file or an introspection query to discover the available types, queries, and mutations.

query-graphql

Executes GraphQL queries against the configured endpoint, enabling you to fetch data or perform operations through the MCP interface.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational