CodeTwin

Your Machine. Your Rules.

7 followers

Your Machine. Your Rules.

7 followers

Entirely on your machine, builds a pre-flight impact map before changes, and gives you five autonomy levels to match your risk tolerance. Your code stays local, you bring your own LLM key, and twin memory lives in a local SQLite store. The launch includes the CLI/TUI, the mobile remote control app, and the bridge that connects them

Free Options

Launch tags:Productivity•Artificial Intelligence•GitHub

Launch Team

LightfieldAI-native CRM that builds itself and does work for you

Promoted

Maker

📌

Hey Product Hunt! Today we’re launching CodeTwin - a terminal‑first AI coding agent that runs locally, with a companion mobile app so you can approve, steer, and monitor runs from anywhere. I built it because existing agents felt either too opaque or too cloud‑dependent; I wanted speed without losing control. During the build, we kept tightening the trust loop: pre-flight impact maps before edits, five autonomy levels you can change mid‑task, and local twin memory (SQLite) so your context stays on your machine. BYOK, zero telemetry, no vendor lock‑in. Would love your feedback: what’s the #1 thing you wish local‑first coding agents did better?

Report

4mo ago