just-every/mcp-read-website-fast

Overview

An MCP server that reads webpages and converts them to markdown, designed for fast and token-efficient web scraping. It is built in TypeScript and provides a simple interface for AI agents to consume web content with minimal token usage.

Best for

Best for
Developers building AI agents that need efficient, low-token web content extraction

Use cases

• Scraping documentation or articles for LLM context
• Extracting clean text from webpages for analysis
• Feeding web content into AI agents with low token overhead

How to use

Install

claude mcp add read-website-fast -s user -- npx -y @just-every/mcp-read-website-fast

Tools exposed

• read_website

Tested with

Claude Code, VS Code, Cursor, JetBrains IDEs

Example client config

{\n  "mcpServers": {\n    "read-website-fast": {\n      "command": "npx",\n      "args": ["-y", "@just-every/mcp-read-website-fast"]\n    }\n  }\n}

Notes

An MCP server that reads webpages and converts them to markdown, designed for fast and token-efficient web scraping. It is built in TypeScript and provides a simple interface for AI agents to consume web content with minimal token usage.

150 stars on GitHub. Last updated 2026-05-01. Licensed MIT.

Use cases

• Scraping documentation or articles for LLM context
• Extracting clean text from webpages for analysis
• Feeding web content into AI agents with low token overhead

Pros

• Fast conversion to markdown reduces token consumption
• Simple, focused tool with a clear purpose
• Open source with 150 stars, indicating community interest

Cons

• May not handle JavaScript-heavy or dynamic pages well
• Output is limited to markdown, no structured data extraction
• Requires an MCP-compatible client to use

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