scrapedo/scrapedo-scraper-tool

Web Page Scraper — Powered by Scrape.do

Extract any public web page as clean Markdown, ready to feed into LLMs.

Powered by Scrape.do:

• —Bypasses anti-bot systems, WAFs and CAPTCHAs (Cloudflare, PerimeterX, DataDome, Akamai)
• —110M+ rotating datacenter, residential, and mobile proxies across 150 countries
• —Optional JavaScript rendering for SPAs (React, Vue, Angular)
• —Built-in conversion to LLM-friendly Markdown output

🔑 Get your free API token

👉 [Sign up free on Scrape.do](https://dashboard.scrape.do/sign-up?utm_source=huggingface&utm_medium=scraper) — 1,000 successful requests per month, no credit card required.

Paste your token into the UI and start scraping.

Use as a smolagents Tool

from smolagents import CodeAgent, InferenceClientModel
from tool import ScrapeDoScraperTool

agent = CodeAgent(
    tools=[ScrapeDoScraperTool(api_token="YOUR_SCRAPEDO_TOKEN")],
    model=InferenceClientModel(),
)
agent.run("Summarize the front page of https://news.ycombinator.com")

You can also set SCRAPE_DO_API_TOKEN in your environment and instantiate ScrapeDoScraperTool() without arguments.

Inputs

• —url (required) — Target URL to scrape.
• —render (optional) — true to run a headless browser for JS-heavy pages.
• —super_proxy (optional) — true to route through residential/mobile proxies for tough anti-bot targets.
• —geo_code (optional) — ISO country code (e.g. us, gb, de, tr) for geo-targeted requests.

Learn more at scrape.do · Documentation

Check out the configuration reference at <https://huggingface.co/docs/hub/spaces-config-reference>.