webclaw

webclaw is a local-first Rust web extraction tool for turning pages into markdown, JSON, or LLM-ready context, with CLI, API, SDK, and MCP entry points. Why it matters: Cleaner source capture can improve research agents, RAG ingestion, and monitoring workflows by reducing noisy page chrome before model processing.

Try this: Compare its markdown output against one existing web-capture method on three documentation or changelog pages, then inspect whether citations and structure survive cleanly.