Sled - Run your coding agent from your phone, with voice

Sled lets you run your coding agent from your phone using voice. Coding agents need frequent input, but when you step away from your desk they just sit idle. Sled solves this by giving you a voice interface that connects securely to your local agent over Tailscale. Your code never leaves your machine. You talk, the agent works locally, and you hear the result read back. Works with Claude Code, OpenAI Codex, and Gemini CLI. Fully open source.

Add a comment

Replies

Best
Hey Product Hunt 👋 We built Sled because our coding agents kept getting stuck whenever we stepped away from our desks. They need input every 10–60 minutes — and terminals are terrible for that when you’re not sitting in front of them. Sled gives you a voice interface to your local coding agent. You talk from your phone, the agent runs on your computer, and you hear what it did. No code leaves your machine — everything goes over Tailscale. It’s fully open source and takes about 5 minutes to set up. We’d love feedback, questions, or ideas for where this could go next 🛷

 Hi Jack, congrats on the launch. This is brilliant in theory; tell me how you've solved the silly practical problems; driving/transit, looking up architecture/features and other deskbound stuff?

 Thanks! Good questions! I would say this is a supplement to desk work not a complete replacement. And of course I don't recommend driving with it but in theory it should be good anywhere it's safe to speak with voice.

 Nice. Congrats on placing!

 congrats!

what happens if the speech to text hallucinates a command? does it have a dry run mode or confirmation step for high risk shell commands?

   voice messages don’t auto send, so you can review transcribed text before sending it.

  Congrats on the launch! Is it connected to Layercode or those are two different products?

Congrats on the launch 👏 Curious when users land on the page, what’s the one action you most want them to take right now?

 this is cool. What about situations where it's not super practical to use voice (on the bus, in a restaurant, etc)?

What happens if the agent needs code review or visual context?

 for visual context, you can connect a browser skill and trust the agent to make a good choice. It's more for communicating high level tasks on the basis that you already trust it to do good stuff.

Really awesome team! Super excited to try it out

 Thanks mate, hope you're well!

 I'm great. Really happy to see you topping the charts

solving the need to babysit local agents every 10–60 minutes is such a smart wedge. is this agnostic to the agent framework I'm using, or is it optimized for specific ones like OpenDevin?

Well done, congrats on the launch!

 Thanks so much Marek!

Congrats on the launch - always appreciate ways to enhance vibe coding productivity, very cool!

What are other future planned integrations down the line? Cursor (fingers crossed)?

 Thanks! Yes we'd love to add Cursor support soon!

Congrats on the launch.What about reviewing the code state

 right now it's focused on just telling the agent to do stuff and trusting them (though you can stop their activity completely). And less focused on reviews for this launch. But based on all the requests we should build in a great review flow!

what a great idea , I'm pretty sure it can save so much time! congrats on the launch 👏

 Thanks!

Congrats on the launch! Love how Sled unlocks hands‑free coding by securely bridging mobile voice input with local agents over Tailscale—super compelling for Claude Code power users.​

 Thanks!

 This solves a real agent problem, not a UX gimmick.

Local coding agents don’t stall because they’re dumb — they stall because they need frequent human steering. Sled keeps execution local and uses voice as a lightweight control plane, which is exactly the right abstraction.

Tailscale + no code leaving the machine is a solid design choice. Feels like missing infrastructure for long-running agent workflows.

Cool idea ,voice control for this is interesting. I noticed one step in the first-run flow where users might hesitate. Happy to share a concrete example if that helps.

 yes please!!

 Great. On first run, the step where voice permissions are requested can feel a bit abrupt. A short line explaining what the user will immediately be able to do with voice (e.g. one example command) could reduce hesitation.

123
Next
Last