Menso
Select supported macOS actions during a voice-driven task.
What Jev decides
Which candidate action to take, or whether work is complete or blocked.
Menso combines a native Mac task runner, live voice input, and Jev selection of exact candidate actions from fresh app context.
- Jev role
- Select the next allowed Mac action from fresh local context.
- Input
- The original task, known apps and controls, and verified completed steps.
- Output
- A candidate ID or a completion or clarification signal.
- Decision frequency
- After each verified task step while the task is active.
- Fallback
- Local approvals and verification gate actions; unsupported tasks stop or ask for clarification.
How it works
Overview
The macOS assistant uses a live voice model to receive and report on a task, while Jev selects the next action from known apps, windows, and controls. The Mac enforces local policy, approvals, execution, and verification. Documented actions currently cover opening apps, focusing windows, inserting exact text without submitting, and setting a focused control.
- Language
- Swift
- License
- MIT
- Author
- Ayush0054
Sources
Related projects
typesafe-computer-use
A native macOS computer-use loop built from local OCR, accessibility data, and bounded Jev actions.
What Jev decides
Which computer action, visible item, site, or off-screen accessibility control to use next.
Jev Chat Assistant
An Android conversation copilot that uses Jev to interpret chats and rank reply candidates without sending for the user.
What Jev decides
The likely intent, conversation risk, reply need, best next action, unresolved tension, and which of three candidate replies fits best.
Distill
Route coding-agent work, reasoning effort, and context retention through bounded Jev decisions.
What Jev decides
Which configured model and effort should handle a call, what evidence deserves attention, whether another attempt is useful, and what context should be retained.