mcp-graphql-tools
GraphQL MCP server for AI assistants
claude mcp add --transport stdio saewoohan-mcp-graphql-tools npx -y mcp-graphql-tools --endpoint=https://api.github.com/graphql --headers={"Authorization":"Bearer YOUR_GITHUB_TOKEN"} --timeout=30000 --maxComplexity=100How to use
This MCP server exposes GraphQL tooling through two main tools: graphql_query and graphql_introspect. graphql_query lets you execute arbitrary GraphQL queries against a specified endpoint, with optional variables, custom headers, a request timeout, and a flag to allow or disallow mutations. graphql_introspect enables you to fetch and explore the GraphQL schema from a given endpoint, with options to provide headers and to include deprecated types and fields in the results. These tools are designed to help AI assistants interact with GraphQL APIs in a structured, automated way by standardizing input parameters and responses.
To use these tools, run the server via NPX as shown in the example configuration. Once running, you can invoke graphql_query by supplying a query string, an optional variables object, a target endpoint, optional headers, and other controls like timeout and allowMutations. For schema exploration, use graphql_introspect with the endpoint and headers to retrieve the schema, then iteratively inspect types, fields, and deprecations to guide subsequent queries.
How to install
Prerequisites:\n- Node.js and npm or npx installed on your system. You may verify with node -v and npm -v.\n\nInstallation steps:\n1) Ensure you have access to the npm registry (or a network path that can reach npm).\n2) You do not need to install a global package for this MCP server; it is intended to be run via NPX as shown in the configuration. If you want to install the underlying tool locally, you can run:\n\n npm install -g mcp-graphql-tools\n\n3) Run the MCP server using the configuration example: use NPX with the provided arguments, for example:\n\n npx -y mcp-graphql-tools --endpoint=https://api.github.com/graphql --headers={"Authorization":"Bearer YOUR_GITHUB_TOKEN"} --timeout=30000 --maxComplexity=100\n\n4) Ensure you replace YOUR_GITHUB_TOKEN with a valid token or adjust headers as needed.\n\nPrerequisites recap: Node.js environment, network access to the GraphQL endpoint, and any required authentication tokens for the target endpoint.
Additional notes
Tips and notes: - The endpoint and headers are configurable per invocation; the defaults shown are examples (GitHub GraphQL API). - Headers must be provided as a JSON string; ensure proper escaping in your shell or JSON configuration. - The maxComplexity option helps prevent expensive queries; tune it based on your API capacity. - If using introspection, you may want to include deprecated types to understand legacy parts of the schema. - When using NPX, you’re running the package without a global install, which is convenient for ephemeral or CI usage. - If you encounter network or authentication errors, verify token validity, endpoint accessibility, and firewall rules.
Related MCP Servers
mcp-graphql
Model Context Protocol server for GraphQL
git
An MCP (Model Context Protocol) server enabling LLMs and AI agents to interact with Git repositories. Provides tools for comprehensive Git operations including clone, commit, branch, diff, log, status, push, pull, merge, rebase, worktree, tag management, and more, via the MCP standard. STDIO & HTTP.
mcp-ts-template
TypeScript template for building Model Context Protocol (MCP) servers. Ships with declarative tools/resources, pluggable auth, multi-backend storage, OpenTelemetry observability, and first-class support for both local and edge (Cloudflare Workers) runtimes.
time
⏰ Time MCP Server: Giving LLMs Time Awareness Capabilities
awesome-remote s
A curated list of Hosted & Managed Model Context Protocol (MCP) Servers accessible via a simple URL endpoint.
ContextPods
Model Context Protocol management suite/factory. An MCP that can generate and manage other local MCPs in multiple languages. Uses the official SDKs for code gen.