Turnal
A flight recorder for AI coding agents.
11 followers
A flight recorder for AI coding agents.
11 followers
Turnal is a local-first flight recorder for AI coding agents. It captures what an agent changed, why it changed it, and the state of your workspace before and after every turn so you can inspect, understand, and roll back agent work with confidence.
Free
Launch Team
AWS Startup AdvisorPersonalized AWS guidance that meets you where you’re already building
Promoted
Hey Product Hunt! I’m excited to launch Turnal, a local accountability layer for AI coding agents.
Website: https://turnal.johari-dev.com/
AI agents are becoming capable of making large, multi-file changes across a codebase. But when something goes wrong, it can still be difficult to answer basic questions:
What exactly did the agent change?
Why did it make that decision?
What did the project look like before that turn?
How can I undo one bad step without losing everything else?
Turnal records each agent interaction as an auditable checkpoint. It captures the agent’s activity alongside snapshots of the workspace before and after every turn.
Turnal is designed to be local-first. Your source code and activity history stay on your machine, and normal recording does not add commits to your existing Git history.
The goal is not to slow agents down or require approval for every edit. It is to let agents
move quickly while making their work observable, reversible, and accountable.
Turnal is publicly available starting today. I’d love to hear how you currently review, debug, or roll back work performed by coding agents, and what you think an agent accountability layer should capture next.