Hands-on Use Cases
Choose what you want to accomplish. Each guide includes a starting state, numbered actions in the actual Control UI, expected evidence, and recovery steps. The guides are for the development implementation indatarelay-labs/datarelay-control main-v2 @ f7ae7b35 (18 October 2026), not a blanket assertion that every feature has passed release or browser acceptance.
Prefer visuals? Follow the numbered Stream Wizard screenshots, from Connector selection to deployment and Logs.
Connector-first workflow
① Review the vendor API guide → ② Prepare access/auth/sample data → ③ Ask ChatGPT with public documentation and redacted samples if uncertain → ④ Create and test Connector → ⑤ Create Destination → ⑥ Create Stream and confirm records/checkpoint → ⑦ Mapping, masking and supported tokenization → ⑧ Deploy → ⑨ Verify Logs and real receiver → ⑩ Troubleshoot delivery/schema drift. Each Connector guide now starts with its specific inputs, an actionable ChatGPT prompt, the full UI workflow and failure diagnostics. Screenshots must be authentic and version-verified; do not treat placeholders as product evidence.Start here
- Prepare a safe lab — a runnable Python HTTP receiver, test payload, and connectivity checks.
- Webhook → first delivered event — create Connector, Destination, Stream, Route, and verify the receiver.
- HTTP API collection — configure a real API Connector, sample records, and checkpoint behavior.
More source and destination integrations
Delivery and processing scenarios
Mental model
Evidence checklist for every scenario
- Connection or Source test shows the expected sample or an explicitly recorded limitation.
- Deploy shows readiness; warnings are reviewed, not silently ignored.
- Stream is active and a new run is visible.
- Each expected Route has a matching delivery log and a receiver-side record (when delivery applies).
- A failed, quarantined, or blocked event is not counted as successful delivery.