Lystn speaks your AI coding assistant's replies out loud so you can pace, refill coffee, or watch another monitor and still follow exactly what Claude Code or Codex just did. It summarizes the wall of text down to what matters, and your code never leaves your machine.

Hey Product Hunt 👋 I'm Burak, the maker of Lystn. I build things with Claude Code and Codex all day, and I kept hitting the same wall: the assistant finishes, dumps a wall of text, and I have to stop everything and read it to know what it did. As someone who loses focus easily, that constant context-switch killed my flow. So I built Lystn — it speaks your AI coding assistant's replies out loud. It summarizes the long ones down to the sentences that matter, gives each terminal its own voice (so you can run several agents and always know who's talking), and lets you just listen while you keep moving. Two things I care about: • Privacy — your code, file paths, and links are stripped on your machine before anything is sent. Your actual code never leaves your device. • Zero friction — one command: npm install -g lystn-cli, then lystn wire. macOS, Windows, Linux. There's a free tier, no card needed. It works with Claude Code and Codex today (Cursor + others next), in 7 languages. I'd genuinely love your feedback — what would make this a daily tool for you? I'm here all day. 🙏

does the summary still capture multi-file diffs reliably when the assistant refactors across like 12 files, or does it start losing the thread on bigger changes?

Been using AI coding tools daily, and this solves a problem I didn't realize bothered me this much. Getting spoken summaries instead of constantly switching back to the terminal is such a smart idea. Congrats on the launch! 🚀

does the summary still capture multi-file diffs reliably when the assistant refactors across like 12 files, or does it start losing the thread on bigger changes?

Lystn

Listen to your AI coding assistant — don't read it

Listen to your AI coding assistant — don't read it