Show HN: Fouine – local full-text and semantic search for scanned PDFs on macOS

Category: other

Tags: macos, full-text-search, semantic-search, ocr, pdf, privacy, swift, mcp

Score: 7.0/10 (Innovation: 6, Technical: 7, Documentation: 9, Utility: 6)

Fouine is a local-first full-text and semantic search engine for macOS that indexes scanned PDFs and many other document formats, using Apple's Vision OCR to extract text from pages without a text layer. It combines hybrid search (fuzzy + embeddings), a command-line tool, and an MCP server for assistant integration, all while keeping documents offline. The project stands out for its thorough documentation, privacy engineering, and practical focus on a real macOS Spotlight gap.

Target audience: macOS users, researchers, and developers needing private document search over scanned and mixed-format archives

Repository: https://github.com/basedpolymer/fouine · Swift · NOASSERTION

View on Hacker News