WebShift
Denoised web search — fetch, clean, and rerank web content for AI agents with strict size budgets.
What it does
Denoised web search — fetch, clean, and rerank web content for AI agents with strict size budgets.
WebShift is a Rust library and MCP server that transforms noisy HTML web pages into clean, right-sized plain text optimized for LLM consumption. Features streaming HTTP fetching with per-page size caps, HTML noise removal, Unicode/BiDi sterilization, BM25 deterministic reranking, and adaptive budget allocation across multiple search backends including SearXNG, Brave, Tavily, Exa, Google, and Bing. Enforces hard guarantees on output size to prevent context window flooding.
Capabilities
Server
Quality
deterministic score 0.56 from registry signals: · indexed on pulsemcp · has source repo · 5 github stars · registry-generated description present