My next Astra experiment: create a mini diorama motocross simulator.
Set riders skill levels and sculpt the course.
Used Astra Medium/High depending on tasks. Burned through 70% of my weekly limit already 😆.
Play Mini Motto for free: https://t.co/Th5Ni3J906
Instinct just admitted it leaked someone else's data to me
It described a financial document with details that weren’t mine, then claimed someone else’s photo had crossed into my chat.
But I never sent a photo
@noahrshinn, was this a data leak or a hallucination?
I wish I had the resources to develop this. It just makes sense to me that this should exist in our near future. I imagine the hands have cameras in them to provide 360 awareness and macro awareness upclose.
I wanted to share an open-source project I built for myself called Sense MCP. It's an MCP server that is activated based on the user's ask and pulls data from a variety of sensors, activities, apps or even your personal self report at any given moment as context for your chat.
You just speak naturally and your AI now has more available information to respond to you than ever before, for a more human and personalized response every time.
Below are all the sources it can pull from:
- Apps & windows: Active app, activity category, window label, sensitivity level, optional redacted window title.
- Workspace & code: Project name/type, Git branch, uncommitted changes and severity, package manager, available scripts, inferred work mode.
- Activity & focus: Input idle time, active/idle/away status, estimated input cadence, configured Focus mode and Do Not Disturb.
- Power & devices: Battery level, power source, low-battery flag, external displays, connected AirPods and Bluetooth input devices.
- Physical environment: Ambient light, microphone loudness, noise category, approximate location based on Wi-Fi.
- Media: Spotify or Apple Music playback state.
- Time & schedule: Local time, day segment, weekday/weekend, estimated daylight, current/upcoming calendar events, rough time pressure, work window and preparation urgency.
- Imported health data: Readiness score/category, recovery, stress and sleep debt.
- Imported weather data: Weather conditions, temperature, precipitation, wind, UV and daylight.
- iPhone self-reports: Feeling, energy, stress, focus, typed/dictated notes, context presets, tags and expiry.
- iPhone device & movement: Battery, charging, Low Power Mode, thermal state, device/iOS version, movement type, steps, distance and floors climbed.
- iPhone health & sound: Active calories, recent heart rate, resting heart rate, sleep duration, noise level and category.
- Explicit visual captures: One camera image, one app-window screenshot, or the main display.
- Derived summaries: Suggested interruption level, situation summary, recent changes, confidence, unknowns, risks and signal freshness.
open-source: https://t.co/qFPJGKgkt2