Hello Product Hunt! 👋 I’m Noah, the creator of VibeContext. Like many of you, I’ve fully embraced coding with AI with Claude 3.5 or other LLM. But I quickly hit a wall: I was wasting way too much time cherry-picking files and, more importantly, burning through thousands of tokens by sending "noise" (lockfiles, dist folders, etc.) that confused the AI. The Problem: Existing tools are either CLI-heavy (hard to setup) or Cloud-based (risky for your source code). The VibeContext Solution: I built this tool with a "Browser-Only" approach using the File System Access API. 0% Cloud: Your files never leave your machine. It’s 100% local, 100% private. Token Efficiency: It generates a clean ASCII tree and strips out the junk so the AI understands your project architecture instantly without re-scanning everything. Better context = smarter answers + lower costs. Zero Setup: No npm install, no account required. Just open the URL, select your folder, and get back to coding. I’m launching this as an independent dev, and I’d love to get your feedback! I'll be rolling out a page soon so you can sign up and get notified of the launch before the actual launch on April 22! I'll be here all day to answer your questions! 👇

VibeContext

Transform any local folder into LLM-ready Markdown

Transform any local folder into LLM-ready Markdown