Skip to content
Skryp

Collections that stay current

Collect finds the data source behind a listing page (the site's own API or data embedded in the page), maps the fields you asked for once and checks them against the page. A dataset keeps those records with stable IDs: each refresh reports what was added, changed or removed, and a refresh that fails or stops early never counts as a deletion.

  • Fields in plain words: “model, RAM, price, stock, URL”.
  • Saved as a recipe, so refreshes need no browser and no model.
  • Self-repair: when a site changes, a new method is promoted only if it reproduces saved examples.
  • Export CSV or JSON; schedule refreshes with a webhook for changes.

Limits

  • Works where the records come from a replayable source; otherwise Skryp says so and suggests extraction.

Price

Collect: each page read at its route's price, plus 5 when a model maps the fields. A refresh of a saved recipe reads its pages again with no model. Full rate card

Start freeRead the quickstart
Collect laptops · REST
curl -X POST https://api.skryp.dev/v1/collect \
  -H "Authorization: Bearer $SKRYP_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "url": "https://shop.example/laptops",
  "fields": [
    "model",
    "RAM",
    "price",
    "stock",
    "url"
  ]
}'
Collect laptops · MCP tool call
{
  "tool": "collect",
  "arguments": {
    "url": "https://shop.example/laptops",
    "fields": [
      "model",
      "RAM",
      "price",
      "stock",
      "url"
    ]
  }
}

Related