leonardcser

Author here, thanks! The selling point is really compiled performance, no runtime, <50mb binary and self contained.

You can use the following to move to prev/next prompt: Alt+Shift+Up/Down or [u and ]u or you can customize it:

smelt.keymap.set("", "<C-Up>", function() return smelt.transcript.move_block("u", { forward = false }) end, { desc = "previous user message" })

smelt.keymap.set("", "<C-Down>", function() return smelt.transcript.move_block("u", { forward = true }) end, { desc = "next user message" })

Hey HN! I've been building custom dev tools with AI agents for the past year and finally wrote up the journey.

Would love to hear how others are approaching their workflows.

Thanks for the feedback. The main performance focus was rendering.

Claude Code and other TUIs (except Codex) use a layer of abstraction over the raw terminal escape sequences.

I directly used `crossterm`, which gave me more control and lower latency.

For example if nothing is going on, I don't render anything in the terminal. Or only render at keypress.

Hello HN

Claude Code is great, but I find it a bit laggy sometimes. I built a local alternative that's significantly more responsive with local models. Just wanted to share :)