ja

jae-jae/fetcher-mcp

๐Ÿ“‡ ๐Ÿ  MCP server for fetching web page content using Playwright headless browser, supporting Javascript rendering and intelligent content extraction, and outputting Markdown or HTML format.

#web scraping#Playwright#content extraction
Publisherjae-jae/fetcher-mcp
Submitted date4/19/2025

Overview

Title

Fetcher MCP

How to Use

  • Run directly using npx -y fetcher-mcp
  • Install required browser with npx playwright install chromium
  • Enable debug mode with --debug flag
  • Configure in Claude Desktop via claude_desktop_config.json

Key Features

  • JavaScript Support: Executes dynamic content via Playwright.
  • Intelligent Extraction: Uses Readability algorithm to extract main content.
  • Flexible Output: Supports HTML and Markdown formats.
  • Parallel Processing: Batch-fetch multiple URLs with fetch_urls.
  • Resource Optimization: Blocks unnecessary resources for efficiency.
  • Configurable Parameters: Customize timeouts, content extraction, and output.

Use Cases

  • Scraping dynamic web pages with JavaScript rendering.
  • Batch processing multiple URLs efficiently.
  • Extracting clean content (removing ads, navigation).
  • Debugging and manual authentication via debug mode.
  • Integration with Claude Desktop for enhanced web fetching.

Visit More

View All