NodeFox

HTTP integrations

Connect a workflow to an existing system

Use a Reader → API operation to retrieve records and a Writer → API operation to send a prepared update. Configure the endpoint, method, headers, credential reference and response or body format for the system you are connecting.

A typical operational flow is:

Reader API → Code (normalize) → Decision (check) → Writer API

For example, retrieve an order and shipment status, compare their identifiers and quantities, and send an approved status update. Each API must support the operation and credentials you configure; a workflow does not create access to an application automatically.

Define the contract

  • Map input and output fields explicitly, including missing or null values.
  • Use stable record identifiers and correlation IDs across systems.
  • Apply the receiving system's authentication and rate limits.
  • Handle expired credentials, unavailable records and rejected writes.
  • Use idempotency keys or a reconciliation check where a repeated request could duplicate an action.

Browser requests are subject to the destination's CORS policy. Local services, supported gateway operations and headless execution have different network boundaries; choose the execution environment before committing to an integration design.

Receive an external trigger

The headless webhook listener accepts authenticated triggers on a customer-controlled host. It is separate from the browser editor and from outbound Reader/Writer HTTP requests.

If your application needs a persistent job-status API, tenant authorization or a business-facing response format, implement that boundary around the runner. There is no hosted NodeFox /v1/workflows service implied by these examples.

Continue