openapi
Allow AI to wade through complex OpenAPIs using Simple Language
claude mcp add --transport stdio janwilmake-openapi-mcp-server wrangler dev
How to use
This MCP server provides an interface to search and explore OpenAPI specifications via oapis.org. It follows a three-step process: determine the relevant OpenAPI identifier, generate a plain-language summary of that spec, and then identify specific endpoints and explain how they work in simple terms. The server supports both JSON and YAML OpenAPI formats and is designed to work with Claude/Cursor style interactions to help you navigate API operations, status codes, and request/response patterns. You can explore an OpenAPI spec at a high level (overview), drill down into individual operations, and learn how to perform common tasks such as authentication, parameter usage, and error handling through concise natural-language explanations. For local testing, you can run the server with Wrangler and then use the MCP inspector to validate and interact with the model through the inspector tool.
How to install
Prerequisites:
- Node.js and npm installed on your system
- Wrangler CLI installed (for Cloudflare Workers development)
- Access to Cloudflare account for Wrangler configuration (optional for local testing)
Install steps:
-
Install Wrangler globally if you haven’t already: npm install -g wrangler
-
Clone the MCP server repository or download the source code: git clone https://github.com/janwilmake/openapi-mcp-server.git cd openapi-mcp-server
-
Install dependencies (if package.json is present): npm install
-
Configure Wrangler for local development (authenticate if needed): wrangler login
-
Run the server locally for testing: wrangler dev
-
Optionally, run the MCP inspector in another terminal to interact with the server: npx @modelcontextprotocol/inspector
Additional notes
Notes:
- The server is intended to work with OpenAPI specs provided by oapis.org and supports both JSON and YAML formats.
- Local testing uses Wrangler to emulate Cloudflare Workers. The MCP inspector can be used to validate prompts and inspect the model’s behavior.
- If you deploy, you may need to adjust environment variables or bindings required by your deployment target (e.g., API endpoints, authentication tokens, or OpenAPI source URLs).
- As you customize the server, you can extend capabilities to filter endpoints, summarize specific operations, or provide additional guidance for complex parameters.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud