This MCP server converts HTML to Markdown or extracts plain text for web-scraping agents. It provides tools for Turndown-based conversion and for stripping an HTML page to whitespace-normalized text, supporting downstream read-the-page workflows that require consistent formatting.

🛠️ Key Features

• html_to_md(html, ...options): converts HTML to Markdown with Turndown-based defaults (ATX headings, fenced code, inline links)
• Strips scripts/styles by default during conversion
• extract_text(html): strips everything and returns whitespace-normalized text
• Read-the-page and web-scraping-focused tooling

🚀 Use Cases

• Generate clean Markdown from HTML for scraping agents
• Extract only text content from a web page when formatting is unnecessary

⚡ Developer Benefits

• Consistent Markdown output (ATX headings, fenced code, inline links)
• Faster path for text-only extraction via extract_text
• Practical defaults to reduce per-turn formatting inconsistency

⚠️ Limitations

• Designed specifically for HTML → Markdown conversion and text stripping (not general document transformation)