// what it does

Mutter sits in your macOS menu bar. Press Cmd+Shift+M, say something, and it figures out what to do with it. No wake words, no cloud. MLX Whisper handles speech-to-text locally, an LLM in LM Studio routes your speech into tasks, notes, or queries against your own knowledge base.

There's also a dictate mode (Cmd+Shift+;) that transcribes your speech and puts it straight on your clipboard. No routing, no classification, just voice to text.

// why I built this

I kept losing thoughts mid-coding-session. The friction of opening a todo app was enough to lose the idea. Mutter removes that: hotkey, talk, done.

Part of a bigger picture: Hive for orchestration, Unplug for defense, Mutter for voice. Each one explores a different facet of making AI work locally.

// stack

Python rumps (menu bar), MLX Whisper (STT), LM Studio (LLM), ChromaDB (vectors), SQLite (tasks), FastAPI (server). CLI: mutter tasks, mutter notes, mutter ask "..."

// what's next

WhatsApp/Telegram delivery, continuous listening, calendar integration, Hive agent integration for autonomous follow-ups, Linux/Windows support.