A tool list spanning dozens of platforms is a large standing cost paid on every single turn, whether or not you scrape anything. Being able to register only the platforms you actually use turns that from a permanent tax into a choice. Set it before you connect, not after. And call get_usage early: knowing what a scrape costs against your quota is the difference between a research session and a surprise.
An aggregator over many scraping targets: search engines, retail marketplaces, social platforms, travel and property listings, job boards, and a generic URL-to-markdown extractor.
- Search — Google organic results, ads and AI Overview
- Retail — Amazon, Walmart, eBay, Target and Home Depot
- Social — YouTube, TikTok, Instagram, Reddit, X and LinkedIn
- Travel and local — Booking, Airbnb, Yelp and Tripadvisor; property via Zillow and Redfin; jobs via Indeed and Glassdoor
- Extract — any URL to markdown or HTML
- get_usage, which they document as free to call
An API key from scavio.dev, as a header. A platforms environment variable controls which targets register, which is how you keep the tool list from filling your context. It also runs locally as an npm package if you prefer.
Paste a URL with a key — add the endpoint to your client