Skip to content

v1.0.0

Latest

Choose a tag to compare

@sergey-ermakovich sergey-ermakovich released this 10 Sep 05:41
· 17 commits to main since this release

First release of the Web Scraping MCP server.

One read-only tool over HasData's hosted endpoint: fetch any public URL through datacenter or residential proxies in 13 countries, with optional JavaScript rendering, wait conditions, page interactions, CSS or AI extraction rules, screenshots and resource blocking. Returns markdown, plain text, raw HTML or structured JSON.

A plain fetch costs 1 credit, so the monthly free tier covers 1,000 of them. Rendering is the main cost lever, and the table is in the README.

  • npm: @hasdata/web-scraping-mcp
  • Endpoint: https://mcp.hasdata.com/api/mcp?apis=web_scraping

This is the fallback for sites with no dedicated tool in the catalogue.

The PyPI package and the MCP registry entry follow shortly. The publish workflow stays disabled until trusted publishing is configured for both packages.