Sled is an open-source tool that enables developers to interact with their local AI coding agents (Claude Code, OpenAI Codex, Gemini CLI) via voice commands from their mobile phones. It securely tunnels voice input to your local machine, processes it, and returns spoken responses, allowing for hands-free coding while away from your desk.
Free
How to use Sled?
Clone the repository, install dependencies, and start the server on your local machine. Set up a secure tunnel using ngrok or Tailscale for mobile access. Connect your phone to the server, and you can start giving voice commands to your coding agent. The agent runs locally, processes your requests, and the responses are converted to speech and sent back to your phone.
Sled 's Core Features
Enables voice-controlled interaction with local AI coding agents from your mobile device, freeing you from your desk.
Provides real-time notifications when your agent completes a task or encounters an issue, keeping you informed.
Offers high-quality voice input and output with support for technical terms like camelCase and over 300 voice options.
Ensures code privacy and security by running the AI agent entirely on your local machine; code never leaves your computer.
Utilizes secure tunneling (ngrok/Tailscale) for safe remote access and is fully open-source for customization and self-hosting.
Sled 's Use Cases
For developers who want to continue coding tasks while away from their desk, such as during a walk or from the couch.
For teams needing to manage or monitor long-running AI coding agent tasks remotely without being tethered to a terminal.
For developers seeking a hands-free coding experience, using voice commands to instruct agents without typing.