@camunda8/hub-api-client

A typed TypeScript client for the Camunda Hub public API v2.

Generated from the Camunda Hub public API v2 OpenAPI contract, so request, response, and model types always match the published spec. Built on openapi-fetch for a small, typed fetch client.

Installation

npm install @camunda8/hub-api-client

Usage

Create a client with the API base URL (including the /api/v2 segment) and a bearer token, then call endpoints by path and method. Request bodies, query parameters, path parameters, and responses are all fully typed from the spec.

import { createHubApiClient } from '@camunda8/hub-api-client';

const client = createHubApiClient({
  baseUrl: 'https://modeler.camunda.io/api/v2',
  // A static token, or a (possibly async) function returning a fresh one.
  token: async () => fetchAccessToken()
});

// Typed path + body. `data` is the typed response, `error` the typed problem detail.
const { data, error } = await client.POST('/projects', {
  body: { name: 'My project', workspaceKey: 'my-workspace-key' }
});

if (error) {
  throw new Error(`Request failed: ${error.detail}`);
}

console.log(data.projectKey);

// Path parameters are typed too. Every call returns `{ data, error }`.
const { data: project } = await client.GET('/projects/{projectKey}', {
  params: { path: { projectKey: data.projectKey } }
});

Authentication

token may be a string or a function returning a string (sync or async). When provided, an Authorization: Bearer <token> header is attached to every request. The function form is resolved on each request, so it can return a freshly refreshed token. Omit token to handle authentication yourself (e.g. via a custom fetch or default headers, both forwarded to openapi-fetch).

Using the model types

The generated models are exported for reuse in your own code:

import type { Schemas } from '@camunda8/hub-api-client';

type Project = Schemas['ProjectResult'];

The lower-level paths, operations, and components types are also exported for advanced use cases.

Contributing

The client is generated from the Camunda Hub public API v2 OpenAPI spec and is not committed. See CONTRIBUTING.md for how to fetch the spec, regenerate, and publish.