AnyAPI

You connect any REST API to AI assistants by using a universal MCP server that exposes every endpoint from an OpenAPI or Postman spec, enabling GraphQL-style field selection, automatic schema inference, and seamless querying without writing custom server code.

How to use

You leverage MCP clients to access any REST API through the universal MCP server. Start by providing the API specification to the server, then discover endpoints, inspect their behavior, and query exactly the fields you need. You can compose queries that fetch multiple endpoints in parallel and override per-request headers when needed.

How to install

Prerequisites: you need Node.js and npm installed on your machine.

Install the MCP server globally using npm.

npm install -g anyapi-mcp-server

Additional sections

Configure and run one or more MCP instances by using the provided CLI examples. You may point each instance at a specific OpenAPI spec or Postman collection and set a base URL for the target API. You can also supply per-request headers and enable request logging for auditing.

Below are concrete startup configurations for commonly used APIs. Each configuration runs the MCP server locally via a command similar to npx and passes the necessary flags and arguments.

{
  "mcpServers": {
    "cloudflare": {
      "command": "npx",
      "args": [
        "-y",
        "anyapi-mcp-server",
        "--name", "cloudflare",
        "--spec", "/path/to/cloudflare-openapi.json",
        "--base-url", "https://api.cloudflare.com/client/v4",
        "--header", "Authorization: Bearer ${CLOUDFLARE_API_TOKEN}"
      ],
      "env": {
        "CLOUDFLARE_API_TOKEN": "your-cloudflare-api-token"
      }
    }
  }
}

{
  "mcpServers": {
    "datadog": {
      "command": "npx",
      "args": [
        "-y",
        "anyapi-mcp-server",
        "--name", "datadog",
        "--spec", "/path/to/datadog-openapi.json",
        "--base-url", "https://api.datadoghq.com/api/v1",
        "--header", "DD-API-KEY: ${DD_API_KEY}",
        "--header", "DD-APPLICATION-KEY: ${DD_APP_KEY}"
      ],
      "env": {
        "DD_API_KEY": "your-datadog-api-key",
        "DD_APP_KEY": "your-datadog-app-key"
      }
    }
  }
}

{
  "mcpServers": {
    "metabase": {
      "command": "npx",
      "args": [
        "-y",
        "anyapi-mcp-server",
        "--name", "metabase",
        "--base-url", "https://your-metabase-instance.com/api",
        "--header", "x-api-key: ${METABASE_API_KEY}"
      ],
      "env": {
        "METABASE_API_KEY": "your-metabase-api-key"
      }
    }
  }
}

Tools and capabilities

The server provides a set of tools to explore and fetch data from REST APIs through a GraphQL-like workflow.

Available tools include the following core actions.

list_api
call_api
query_api
explain_api
batch_query

Available tools

list_api

Browse and search available API endpoints from the spec. Returns categories and endpoint counts to help you navigate the API surface.

call_api

Inspect an API endpoint by making a real request and returning the inferred GraphQL schema. Does not return data, only the schema and suggested queries.

query_api

Fetch data using a GraphQL-like query, returning only the fields you select. Supports pagination and per-request headers.

explain_api

Get detailed endpoint documentation directly from the spec, including parameters, body schema, response codes, and deprecation status without making HTTP requests.

batch_query

Fetch data from multiple endpoints concurrently in a single call, returning results for each request independently.