Sled allows you to run coding agents from your phone, making it easier to interact with them when you're away from your desk. Coding agents typically need input every 10 to 60 minutes, and Sled utilizes voice commands to speed up this process. By running in your browser, it circumvents the limitations of traditional terminals that can’t handle two-way voice communication. You can simply talk to your agent, and Sled transcribes your speech and sends it over a secure connection to your local machine, ensuring that your code never leaves your computer. You can receive notifications on your phone when your agent completes a task or encounters an issue. The responses from the agent are converted to speech, with over 300 voice options available. This hands-free setup allows you to work from various locations—whether you're on the couch, in bed, or out for a walk—without needing to keep your laptop close by. Sled supports secure tunneling through services like ngrok or Tailscale, and it’s an open-source project that you can modify or self-host if desired. Setting up Sled is straightforward, taking about five minutes if you have a compatible coding agent and a tunneling service. The installation process involves cloning the repository and running a few commands. However, be cautious with security. If you expose your machine without proper authentication, you risk giving others control over it. While your audio and agent responses are processed through Layercode, they’re not stored, and you have the option to disable voice output to keep everything local.