Scrapeless Google AI Overview Scraper

Capture Google's AI Overview answer with its sources as structured JSON, in Node.js or Python.

The Scrapeless LLM Chat Scraper reads Google AI Overview's response as data: send a prompt to the scraper.overview actor and get the answer back as structured fields — no browser to drive, no anti-bot to fight.

---

How it works

Two calls, both authenticated with the x-api-token header:

1. Create — POST https://api.scrapeless.com/api/v2/scraper/request with {"actor":"scraper.overview","input":{"prompt":"...","country":"US"}} → returns a task_id.
2. Fetch — GET https://api.scrapeless.com/api/v2/scraper/result/{task_id} → poll until status is success.

Field	Where	Meaning
actor	top level	scraper.overview
input.prompt	input	the question to ask Google AI Overview
input.country	input	ISO country code to localize the answer

---

Quickstart

curl

export SCRAPELESS_API_KEY=sk_...   # free key at https://app.scrapeless.com
curl -X POST https://api.scrapeless.com/api/v2/scraper/request \
  -H "x-api-token: ${SCRAPELESS_API_KEY}" -H "Content-Type: application/json" \
  -d '{"actor":"scraper.overview","input":{"prompt":"What are the top attractions in New York?","country":"US"}}'

Node.js (18+): SCRAPELESS_API_KEY=sk_... node examples/google-ai-overview.mjs Python: pip install -r requirements.txt && SCRAPELESS_API_KEY=sk_... python examples/google-ai-overview.py

---

Examples

File	Language
examples/google-ai-overview.mjs	Node.js
examples/google-ai-overview.py	Python

---

FAQ

Q: Do I need a browser or a proxy? No. The actor renders Google AI Overview server-side through residential egress; you send the prompt and read the JSON back.

Q: Why is the response asynchronous? The task runs server-side and returns a task_id; you poll /result/{task_id} until status is success.

Q: Where are the full docs? docs.scrapeless.com. Part of the Scrapeless LLM Chat Scraper line.

---

License

MIT — see LICENSE. Requires a free API key from app.scrapeless.com.