Create a scraping job
POST a public URL and the desired output formats to /scrape. Synchronous waiting is useful for small jobs; asynchronous delivery is safer for production workloads.
cURL
curl -X POST https://api.nordicdevhouse.com/v1/scrape \
-H "Authorization: Bearer ndh_live_••••••••" \
-H "Content-Type: application/json" \
-d '{
"url": "https://example.com/products",
"formats": ["json"]
}'Job states
| State | Meaning |
|---|---|
| queued | Accepted and waiting for capacity |
| running | Fetching and validating |
| completed | Validated output is available |
| failed | A permanent error stopped the job |
Output contract
Responses include a job ID, stable status, record count, normalized data, and timing metadata. Treat unknown fields as forward-compatible additions.