There's also a shortcut to jump straight into voice mode and it'll auto-detect when you're finished to execute the task.
I made this because I found myself constantly switching between apps and sites to use AI to edit tweets, start documents off, and reword slack messages and wanted something native, fast, and frictionless.
There’s no cloud processing—just direct API calls with your own keys (or no keys, just Ollama models).
I’d love feedback—what do you think? Would this fit into your workflow? Anything missing that would make it more useful?