Structured data extraction

When you know exactly which fields you want, pass a jsonSchema and get back a typed object instead of a page you have to parse yourself.

curl https://request.usestring.ai/v1/fetch \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/product/123",
    "jsonSchema": {
      "type": "object",
      "properties": {
        "name": { "type": "string" },
        "price": { "type": ["number", "null"] },
        "inStock": { "type": "boolean" }
      },
      "required": ["name"]
    }
  }'

Tips

• Mark the fields you truly need in required — if extraction can't fill them, the request falls back to the raw page rather than returning a partial object.
• Add descriptions to your schema properties to guide extraction.
• Keep schemas within the bounds: depth 10, 200 keys, 50 KB.

See Structured extraction for the full rules and the dynamic extraction fee.