cotdp/scraper-mcp repository overview

⁠Scraper MCP

A context-optimized Model Context Protocol (MCP) server for efficient web scraping. Provides AI tools with pre-processed, filtered web content—reducing token usage by 70-90% through server-side CSS filtering and markdown/text conversion.

⁠Quick Start

docker pull cotdp/scraper-mcp:latest
docker run -p 8000:8000 cotdp/scraper-mcp:latest

Access the dashboard at http://localhost:8000⁠

Features

• 4 Scraping Tools: Raw HTML, Markdown, Plain Text, Link Extraction
• CSS Selector Filtering: Extract only relevant content server-side
• Smart Caching: 3-tier cache system with configurable TTLs
• Retry Logic: Exponential backoff for transient failures
• Real-time Dashboard: Monitor requests, cache metrics, and test tools
• Multi-platform: Supports linux/amd64 and linux/arm64

Usage with Docker Compose

services: scraper-mcp: image: cotdp/scraper-mcp:latest ports: - "8000:8000" environment: - TRANSPORT=streamable-http - HOST=0.0.0.0 - PORT=8000 volumes: - cache:/app/cache volumes: cache:

MCP Client Configuration

Connect from Claude Desktop or any MCP client:

{ "mcpServers": { "scraper": { "url": "http://localhost:8000/mcp⁠" } } }

Available Tags

• latest - Latest stable release
• 0.1.0, 0.1, 0 - Semantic version tags
• main- - Latest main branch builds

Documentation

Full documentation: https://github.com/cotdp/scraper-mcp⁠

License

MIT License