SKI

What is SKI

SKI is a free, on-device voice assistant for AI coding agents, available for macOS, Windows, and Linux. It lets you talk out loud to coding tools like Claude Code, Cursor, Codex, and Gemini CLI, and hear their responses in a natural voice. All speech recognition and synthesis happen locally on your machine, ensuring privacy and offline functionality.

How to use SKI

1. Download and install SKI on your Mac (Apple Silicon, macOS 14.4+), Windows (10/11), or Linux (Ubuntu/Debian, x64).
2. Register once (no credit card required) to activate the free version.
3. Launch SKI – it appears as a floating pill (Windows/Linux) or docks into the notch (Mac).
4. Speak to your coding agent – SKI converts your speech to text on-device and sends it to your agent.
5. Hear the response – the agent's reply is spoken aloud in a natural voice and displayed in the pill or notch.
6. Use mute to stop listening, or silent mode to receive text-only responses.
7. Optional: Use the globe/fn key on Mac for push-to-talk or quick mute.

Features of SKI

• Two-way voice loop: Speak to your coding agent and hear it answer out loud.
• On-device processing: Speech-to-text and text-to-speech run locally; nothing is uploaded.
• Offline capability: Works without an internet connection for core voice features.
• Cross-platform: Available for macOS, Windows, and Linux.
• Free forever: Core voice features and meeting transcriber are free with registration.
• Silent mode: Get text-only responses when you don't want audio.
• Mute and push-to-talk: Control when SKI listens.
• Local meeting transcriber: Record and transcribe meetings on-device (free, unlimited).
• AgentCall integration (paid): Send your agent into video meetings as a participant or notetaker.
• Transcripts window: View, edit, copy, export, or delete transcripts and sessions.

Use Cases of SKI

• Hands-free coding: Dictate commands and code changes to your agent while keeping your hands on the keyboard or away from it.
• Accessibility: For developers who prefer voice input or have difficulty typing.
• Quiet environments: Use silent mode to get text responses without audio.
• Meetings: Record and transcribe meetings locally, or send your agent to join as a participant (paid).
• Learning and exploration: Ask your coding agent questions and hear explanations aloud.

Pricing

SKI is free forever on Mac, Windows, and Linux with registration. No credit card is required. The only paid feature is AgentCall, which lets your agent join video calls – billed per minute, with free hours included.

FAQ

What is a "coding agent"? An AI assistant developers run alongside their code, such as Claude Code, Cursor, Codex, or Gemini CLI. It can read files, write code, run tests, and answer questions. SKI gives it ears and a voice.

How does SKI actually work? You speak; SKI turns it into text on your machine and hands it to your coding agent. The agent does the work, then sends its reply back to SKI, which speaks it aloud and shows it in the pill or notch.

Do I need an internet connection? No. Listening, understanding, and speaking all happen on your own machine. The only feature that touches the internet is sending your agent into a video meeting (optional).

Is my voice recorded or sent anywhere? No. Your voice becomes text on your own machine and is never uploaded. There is no cloud service listening in.

How much does it cost? Free forever on Mac, Windows, and Linux – register once, no credit card. Voice features and the meeting transcriber are included. You only pay if you want your agent to join video calls through AgentCall.

What do I need to run it? An Apple Silicon Mac (M1 or later) on macOS 14.4 or newer, a PC on Windows 11 or 10, or a Linux machine (Ubuntu/Debian, x64). English is supported today; more languages are planned.