Skip to content

Lucas/onboarding#13

Merged
rxv801 merged 4 commits into
mainfrom
lucas/onboarding
Jun 12, 2026
Merged

Lucas/onboarding#13
rxv801 merged 4 commits into
mainfrom
lucas/onboarding

Conversation

@lukitasxue

Copy link
Copy Markdown
Collaborator

No description provided.

lukitasxue and others added 4 commits June 3, 2026 18:46
…and removed old assets and components. The new onboarding steps include Camera Setup, Distraction Options, Focus Environment, and Welcome. The styles for the onboarding flow have been updated to enhance the visual appeal and user experience. Old assets such as hero.png, react.svg, and vite.svg have been removed to streamline the design. Additionally, the Onboarding component has been deleted in favor of a more modular approach with individual components for each onboarding step.
* Add claude GitHub actions 1778132378878 (#4)

* "Claude PR Assistant workflow"

* "Claude Code Review workflow"

* feat(cv): phone detection (YOLOX-S via onnxruntime) (#5)

* feat(cv): add phone detector placeholder + detection loop

- phone_detector.py: detect_phone(frame) placeholder, returns protocol-shaped
  event ({type,status,confidence,timestamp}); real detector slots in behind it
- detection_loop.py: camera -> detect_phone -> print loop at 10fps, clean shutdown
- requirements.txt: pin CV/server deps (Python 3.11; MediaPipe lacks 3.13/3.14 wheels)
- setup.sh: one-shot install for Python venv + Electron npm
- docs: python/README.md (incl. camera vs loop separation rationale), root README setup

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* feat(cv): implement phone detection with YOLOX via onnxruntime

Replace the placeholder detect_phone() with a real detector:
- find_phones(frame): YOLOX-S inference -> decoded, thresholded, NMS'd boxes
- detect_phone(frame): per-frame protocol event (status/confidence/timestamp)
- phone_detect_test.py: visual webcam test that reuses find_phones()
- lazy-loaded onnxruntime session (load model once, reuse every frame)
- setup.sh fetches yolox_s.onnx; onnxruntime added to requirements
- gitignore *.task for the upcoming gaze model

Detector is perception only ("phone in this frame?"); distracted-state
policy (smoothing / N-second timer) stays in the loop layer.

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

* whitelist common windows apps detection, and UI, preparing for navigator tabs detection UI and detection logic (possibly)

* minor fixes on build, unused import

* ci: add Electron workflow (#7)

* ci: add Electron workflow

* ci: add Electron and Python workflows

* Lucas/onboarding (#6)

* Onboarding flow updates: added new onboarding steps, updated styles, and removed old assets and components. The new onboarding steps include Camera Setup, Distraction Options, Focus Environment, and Welcome. The styles for the onboarding flow have been updated to enhance the visual appeal and user experience. Old assets such as hero.png, react.svg, and vite.svg have been removed to streamline the design. Additionally, the Onboarding component has been deleted in favor of a more modular approach with individual components for each onboarding step.

* changes on onboarding file names and added camera status and camera selection in camera setup step

* no changes

* merge conflicts solved

* Add browser activity onboarding rules UI and logic

* Clean up camera setup lifecycle and comments

* fixed environment focus settings directory

---------

Co-authored-by: Saatvik Sharma <200460255+rxv801@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
@rxv801 rxv801 merged commit 820e3b4 into main Jun 12, 2026
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants