How do I get a long‑term token?
The `YAPI_TOKEN` is a project‑level token found in your YApi project’s Settings page (under “Token” or “Project Token”).

@xiaozhi666666
Basic information
Reusable MCP server for reading YApi interface metadata from a URL, interface ID, or project ID.
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"yapi": {
"command": "npx",
"args": [
"-y",
"yapi-fetch-mcp"
],
"env": {
"YAPI_HOST": "https://yapi.example.com",
"YAPI_TOKEN": "project-token"
}
}
}
}Read one YApi interface by URL or interface ID and return normalized request/response metadata.
Read a YApi project's interface menu. Useful for discovering interface IDs before calling get_interface.
Search/list YApi project interfaces with pagination.
Yapi Fetch Mcp is an MCP server for reading YApi interface metadata. It can fetch interface path, request parameters, request body, and response structure via YApi page URL, interface ID, or project ID. It is intentionally project‑agnostic, pulling structured data while leaving code generation (services, types, models, API clients) to the coding agent.
Install globally with npm install -g yapi-fetch-mcp. Configure environment variables YAPI_HOST and either YAPI_TOKEN (project token) or YAPI_COOKIE (browser cookie). Then add the server to your MCP client (e.g., Codex) using npx -y yapi-fetch-mcp or the global command. Authentication can also be passed directly in tool parameters.
The `YAPI_TOKEN` is a project‑level token found in your YApi project’s Settings page (under “Token” or “Project Token”).
The current version sends the token only as a query parameter. If your access token must go in the `Authorization` header, use `YAPI_COOKIE` instead, or extend the code for your private YApi.
Yes. Set the `YAPI_COOKIE` environment variable with the full cookie string (e.g., `_yapi_token=...; _yapi_uid=...`). This works for YApi instances that require login.
It uses `/api/interface/get`, `/api/interface/list_menu`, and `/api/interface/list` from the standard YApi API.
Yes. Clone the repository, run `npm install && npm build`, then point your MCP client’s `command` to `node /absolute/path/to/build/src/index.js`.

Model Context Protocol Servers
🔮 Platform-neutral Browser Use for AI agents: real Chrome automation with a CLI + SDKs, no lock-in, dead simple.
The Apify MCP server enables your AI agents to extract data from social media, search engines, maps, e-commerce sites, or any other website using thousands of ready-made scrapers, crawlers, and automation tools available on the Apify Store.
MCP server paired with a browser extension that enables AI agents to control the user's browser.
Allow LLMs to control a browser with Browserbase and Stagehand
Comments