What it does

Exposes Kubb's code generation as a tool, so an agent can regenerate a client from a spec instead of hand-editing forty files. Valuable exactly when the schema changed and the types did not. Requires an existing Kubb setup; there is no incremental value for a project that generates by hand.

Install

Terminal

npx -y @kubb/mcp

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": {
    "kubb": {
      "command": "npx",
      "args": [
        "-y",
        "@kubb/mcp"
      ]
    }
  }
}

Use it for

Generating typed clients from an OpenAPI spec inside a refactor

Skip it if

Hand-written SDKs or projects that do not use Kubb

When to use it

Reach for this when the task needs codegen 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

Kubb MCP