Openapi Inspector
@Nonetss
Openapi Inspector について
Stop pasting API docs. Let your LLM dynamically navigate OpenAPI/Swagger specs and extract precise JSON schemas on-demand.
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"none.api": {
"command": "npx",
"args": [
"-y",
"@nonetss/none.api"
],
"env": {
"OPENAPI_URL": "http://localhost:3000/doc"
}
}
}
}ツール
6List all available endpoints. Uses default URL if not provided.
List all endpoints associated with a specific tag.
Get comprehensive info about an endpoint.
Get the detailed JSON schema required for the request body.
Get the JSON schema of the expected response.
List all tags defined in the specification.
概要
What is Openapi Inspector?
Openapi Inspector is a professional OpenAPI explorer server for the Model Context Protocol (MCP). It allows large language models like Claude or Cursor to dynamically explore, understand, and interact with any API that has an OpenAPI/Swagger specification.
How to use Openapi Inspector?
Run the server directly with npx @nonetss/none.api (or install globally with npm install -g @nonetss/none.api). Configure it in your MCP client (e.g., claude_desktop_config.json) with the command npx -y @nonetss/none.api and set the OPENAPI_URL environment variable to point to your OpenAPI spec.
Key features of Openapi Inspector
- List all available endpoints and methods
- Discover functional sections via tag search
- Fetch granular JSON Schemas for requests and responses
- Support for dynamic OpenAPI URLs at runtime
- Optimized for backends following the hono-template structure
Use cases of Openapi Inspector
- Allow an LLM to understand an unfamiliar API without pasting documentation manually
- Enable code generation by retrieving precise request/response schemas
- Reduce context overload by inspecting endpoints on demand instead of mapping all to tools
- Browse and search API operations interactively during conversation
FAQ from Openapi Inspector
How does Openapi Inspector differ from other MCP API servers?
Instead of mapping every endpoint to a separate tool (which can overwhelm the model's context), it acts as an intelligent inspector that lists endpoints, searches by tags, and retrieves detailed information and schemas on demand.
What are the runtime requirements?
Node.js and npm/npx. The server runs as a command-line MCP server and requires an MCP‑compatible client (e.g., Claude Desktop, Cursor).
Where does the API data come from?
It fetches the OpenAPI specification from the URL provided in the OPENAPI_URL environment variable. The URL can be changed dynamically.
Is authentication supported?
The README does not describe any built‑in authentication mechanism. Authentication must be handled by the OpenAPI endpoint or the client configuration separately.
Are there any known limitations?
The server is optimized for backends following the hono-template structure but works with any valid OpenAPI/Swagger specification. No other limitations are mentioned in the README.
「開発者ツール」の他のコンテンツ

Sentry
modelcontextprotocolModel Context Protocol Servers
Deepwiki MCP Server
regenrek📖 MCP server for fetch deepwiki.com and get latest knowledge in Cursor and other Code Editors
nuxt-mcp / vite-plugin-mcp
antfuMCP server helping models to understand your Vite/Nuxt app better.
Burp Suite MCP Server Extension
PortSwiggerMCP Server for Burp
test
cloudwegoThe ultimate LLM/AI application development framework in Go.
コメント