MCP · Project

Scrapling MCP server

The Scrapling MCP server exposes web page fetching and targeted content extraction from the Scrapling Python package.

+ Follow

Builder facts

  • Package. The self-hosted server comes from the BSD-3-Clause Scrapling package through its optional scrapling[ai] extra. Source ↗
  • Extraction. One-shot tools fetch a URL directly over HTTP or render JavaScript in a browser. CSS selectors can narrow the returned content. Source ↗
  • Sessions. One-shot calls close their own client or browser. Session tools retain browser settings or HTTP cookies across calls. Source ↗
  • Transport. The server supports stdio and Streamable HTTP. HTTP requires a shared token by default and supports accepted-host restrictions. Source ↗

Sources checked

Publisher update ·

Scrapling 0.5.1 adds motion capture to MCP ↗

The MCP browser tools return short recordings as timestamped frames and add mouse hold, drag and release controls.

Change timeline

0 stories

No verified releases in this timeline yet.