mcpField
W
web-scraper

web-scraper

github.com/cowork-osData & databases51761 installsUpdated today
Repository

About web-scraper

A Scrapling-based skill for scraping websites and extracting structured content. It is intended for requests involving dynamic pages or specific data, rather than simple URL fetching or local files.

Key features

  1. 01Scrape website content with Scrapling
  2. 02Target content with a CSS selector
  3. 03Extract text, tables, links, images, and metadata

Use cases

  1. 01Collect structured data from a dynamic web page
  2. 02Extract selected content and links from a URL

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/cowork-os/cowork-os --skill web-scraper

After installing, type/web-scraperin Claude Code to confirm it loaded.

Details

From the project README

Web Scraper

Purpose

Scrape web pages and extract structured data using Scrapling.

Routing

  • Use when: Use when the user wants to scrape content from websites, especially pages with dynamic content or structured data that needs extraction.
  • Do not use when: Don't use for simple URL fetching where web_fetch works fine. Don't use for internal files or local content.
  • Outputs: Extracted web content including text, tables, links, images, and metadata.
  • Success criteria: Successfully extracts the requested content from the target URL with clean formatting.

Trigger Examples

Positive
  • Use the web-scraper skill for this request.
  • Help me with web scraper.
  • Use when the user wants to scrape content from websites, especially pages with dynamic content or structured data that needs extraction.
  • Web Scraper: provide an actionable result.
Negative
  • Don't use for simple URL fetching where web_fetch works fine. Don't use for internal files or local content.
  • Do not use web-scraper for unrelated requests.
  • This request is outside web scraper scope.
  • This is conceptual discussion only; no tool workflow is needed.

Parameters

NameTypeRequiredDescription
urlstringYesURL to scrape
selectorstringNoCSS selector for specific content (optional)

Runtime Prompt

  • Current runtime prompt length: 1117 characters.
  • Runtime prompt is defined directly in ../web-scraper.json.

Comments

Markdown supported