ContextCut PRO: Stop pasting confidential docs into ChatGPT

Hey Product Hunt! 👋 I built ContextCut PRO because I got tired of pasting confidential legal documents into ChatGPT and praying they didn't leak. The idea is simple: a smart proxy sits between your AI tools and Ollama, injects relevant context from your knowledge base into every prompt, and shows you exactly how many tokens you saved. Everything runs locally — your docs, your vectors, your queries never leave your machine. The watcher service is my favorite part: drop a .md file into the knowledge folder and it's searchable in seconds. No re-indexing commands, no config files to edit. I'm here all day to answer questions. What's your biggest frustration with local RAG today? Happy to help you set it up. 🔥