The Thunderbit MCP server is an AI-powered web scraping implementation for the Model Context Protocol (MCP). It can distill web pages into Markdown or extract structured data using JSON Schema, and it is part of a monorepo that also includes a CLI and a Claude Code plugin.
๐ ๏ธ Key Features
- AI-powered web scraping
- Output options: Markdown distillation
- Structured extraction via JSON Schema
๐ Use Cases
- Converting web page content into Markdown for downstream processing
- Extracting fields from web content into structured JSON according to a schema
โก Developer Benefits
- MCP integration for agent workflows
- Schema-driven structured data extraction (JSON Schema)
- Monorepo includes CLI and a Claude Code plugin (in addition to the MCP server)
โ ๏ธ Limitations
- Capabilities described are limited to web scraping, Markdown distillation, and JSON Schema-based structured data extraction.