MCP GraphQL Server

What does “MCP GraphQL” MCP Server do?

MCP GraphQL is a Model Context Protocol (MCP) server designed to provide a standardized interface for interacting with GraphQL APIs. By introspecting a target GraphQL endpoint, it automatically exposes each available GraphQL query as a dedicated MCP tool, enabling AI assistants and developers to seamlessly interact with external GraphQL data sources. This facilitates tasks such as database queries, data retrieval, and integration with third-party services directly through an MCP-compatible workflow. MCP GraphQL handles tool parameter mapping, dynamic JSON schema generation, and authentication (including Bearer, Basic, or custom headers), all without requiring manual schema definition. Its primary goal is to streamline development workflows by making GraphQL APIs accessible and actionable for AI-powered assistants and human users through a unified protocol.

List of Prompts

No prompt templates are mentioned in the provided materials.

List of Resources

No explicit MCP resources are described in the provided documentation.

List of Tools

• GraphQL Query Tools
  Each GraphQL query exposed by the target API is presented as an individual MCP tool. The server dynamically creates a tool for each query, with tool parameters matching the GraphQL query parameters. This allows clients to execute any of the API’s supported queries directly via MCP.
  • Parameters: Automatically generated from the GraphQL schema
  • Input schema: Dynamically built based on the query’s requirements

Use Cases of this MCP Server

• Database Management via GraphQL
  Developers can run complex queries against GraphQL-enabled databases, retrieving structured data directly into their development environment or AI workflows.
• Third-party API Integration
  Seamlessly integrate with SaaS products or services that expose GraphQL endpoints, enabling AI assistants to fetch, aggregate, or manipulate remote data.
• Rapid Prototyping and Data Exploration
  Instantly introspect available queries from a new GraphQL API, allowing for quick experimentation and data analysis without manual schema mapping.
• Automated Reporting
  Use AI agents to automate the extraction of data from GraphQL APIs and compile reports or dashboards on-demand.
• Enhanced File or Content Management
  If a service exposes file or content management capabilities via GraphQL, these can be accessed as tools for streamlined workflow automation.

How to set it up

Windsurf

1. Ensure Python 3.11+ is installed.
2. Install mcp-graphql via pip or use uvx as needed.
3. Edit your Windsurf configuration file.
4. Add the MCP GraphQL server to the mcpServers section:

   "mcpServers": {
     "graphql": {
       "command": "uvx",
       "args": ["mcp-graphql", "--api-url", "https://api.example.com/graphql"]
     }
   }
   

5. Save the config and restart Windsurf.
6. Verify connection by running a test query.
7. Securing API keys: Use environment variables.

   "mcpServers": {
     "graphql": {
       "command": "uvx",
       "args": [
         "mcp-graphql",
         "--api-url", "https://api.example.com/graphql",
         "--auth-token", "${GRAPHQL_TOKEN}"
       ],
       "env": {
         "GRAPHQL_TOKEN": "your-token"
       }
     }
   }
   

Claude

1. Open Claude’s settings/configuration.
2. Install or ensure uvx or mcp-graphql is available.
3. Add to your mcpServers configuration:

   "mcpServers": {
     "graphql": {
       "command": "uvx",
       "args": ["mcp-graphql", "--api-url", "https://api.example.com/graphql"]
     }
   }
   

4. Alternatively, use pip or Docker as shown:

   "mcpServers": {
     "graphql": {
       "command": "python",
       "args": ["-m", "mcp_graphql", "--api-url", "https://api.example.com/graphql"]
     }
   }
   

   or

   "mcpServers": {
     "graphql": {
       "command": "docker",
       "args": ["run", "-i", "--rm", "mcp/graphql", "--api-url", "https://api.example.com/graphql"]
     }
   }
   

5. Save changes and restart Claude to apply.

Cursor

1. Install mcp-graphql using pip or uvx.
2. Open your Cursor configuration file.
3. Add the following to the mcpServers section:

   "mcpServers": {
     "graphql": {
       "command": "uvx",
       "args": ["mcp-graphql", "--api-url", "https://api.example.com/graphql"]
     }
   }
   

4. Save the config and reload Cursor.
5. Test by running a sample query through the MCP interface.

Cline

1. Ensure you have Python 3.11+ and mcp-graphql installed.
2. Locate your Cline configuration file.
3. Add the MCP GraphQL server to your mcpServers settings:

   "mcpServers": {
     "graphql": {
       "command": "uvx",
       "args": ["mcp-graphql", "--api-url", "https://api.example.com/graphql"]
     }
   }
   

4. Save and restart Cline.
5. Use environment variables for tokens as shown above for security.

Securing API Keys

Use environment variables for sensitive data such as tokens:

"mcpServers": {
  "graphql": {
    "command": "uvx",
    "args": [
      "mcp-graphql",
      "--api-url", "https://api.example.com/graphql",
      "--auth-token", "${GRAPHQL_TOKEN}"
    ],
    "env": {
      "GRAPHQL_TOKEN": "your-token"
    }
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "graphql": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “graphql” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

A solid, practical MCP implementation for GraphQL APIs, but lacking explicit prompts/resources and does not mention sampling or roots. Good for tool exposure and ease of setup. Would rate 7/10 based on completeness and utility for developers.

MCP Score