What it does

Reads an OpenAPI document and exposes its endpoints as MCP resources and tools. This is the cheapest path from 'there is an API' to 'the agent can call it', and the least safe: a spec-driven server will happily expose every endpoint, including the destructive ones.

Install

Terminal

npx -y @ivotoby/openapi-mcp-server

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

mcp.json

{
  "mcpServers": {
    "openapi": {
      "command": "npx",
      "args": [
        "-y",
        "@ivotoby/openapi-mcp-server"
      ]
    }
  }
}

Use it for

Turning any documented REST API into agent tools without writing a server

Skip it if

APIs with no OpenAPI spec, or auth flows a spec cannot express

When to use it

Reach for this when the task needs openapi inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
Memory — A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations
Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude

OpenAPI MCP