aparajithn/agent-scraper-mcp

Overview

This MCP server enables AI agents to scrape websites through screenshots, content extraction, and structured data parsing. It exposes scraping capabilities as a Model Context Protocol tool, allowing agents to fetch and process web content during conversations. Built in Python, it uses Playwright or similar browser automation under the hood.

Best for

Best for
Developers building AI agents that need to extract web data programmatically

Use cases

• Extract structured data from websites for AI analysis
• Take screenshots of web pages for visual verification
• Scrape text content to feed into retrieval-augmented generation (RAG) pipelines

Notes

This MCP server enables AI agents to scrape websites through screenshots, content extraction, and structured data parsing. It exposes scraping capabilities as a Model Context Protocol tool, allowing agents to fetch and process web content during conversations. Built in Python, it uses Playwright or similar browser automation under the hood.

4 stars on GitHub. Last updated 2026-03-07. Licensed MIT.

Use cases

• Extract structured data from websites for AI analysis
• Take screenshots of web pages for visual verification
• Scrape text content to feed into retrieval-augmented generation (RAG) pipelines

Pros

• Provides multiple scraping modes (screenshot, text, structured) in one MCP server
• Integrates directly with AI agents that support the Model Context Protocol
• Python-based, easy to extend or modify for custom scraping needs

Cons

• Relatively new project with few community stars, limited documentation or support
• Requires an agent that implements the MCP protocol, which is still emerging
• Dependency on browser automation tools like Playwright adds setup overhead

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.