Mood Compass

Live app: https://moodpredictor.vercel.app

Repository: https://github.com/mikemaeda/moodpredictor

Mood Compass is a privacy-first mood tracking web app built with React, TypeScript, and Vite. It helps users complete quick mood check-ins, track local patterns over time, receive small evidence-informed recommendations, and view gentle personal forecasts once enough history exists.

Mood Compass is intentionally non-diagnostic. It is a wellness and self-reflection tool, not a medical device or replacement for professional care.

Highlights

• Quick mood check-ins with simple state selection
• Optional deeper details: emotions, intensity, sensations, context tags, and notes
• Local-first storage using the browser
• Export and delete controls for user data
• Dashboard with today summary, gentle streaks, weekly review, and mood pixels
• History, insights, tools, settings, privacy, forecasting, and help-now pages
• Evidence-informed recommendation tools
• Transparent forecasting logic with confidence labels
• Accessibility-minded UI with keyboard support, focus states, and reduced-motion support
• Deployed on Vercel

Privacy-First Design

Mood Compass stores each user's data locally in their browser for the MVP. No account is required, and mood history is not sent to a server.

Users can:

• Save mood entries locally
• Export their data as JSON
• Delete all local data
• Disable analytics
• Enable reduced motion

Because storage is local-first, data does not automatically sync across devices or browsers.

How It Works

1. Choose the closest current mood state.
2. Optionally add a primary emotion, tags, sensations, or note.
3. Save the check-in locally.
4. Review trends, patterns, and recommendations over time.
5. After enough history exists, Mood Compass shows a gentle forecast based on personal patterns.

Forecasts are explainable and never presented as certain.

Tech Stack

• React
• TypeScript
• Vite
• CSS
• Local Storage
• Vitest
• Testing Library
• Playwright
• Vercel

Project Structure

src/
|-- components/
|   |-- layout/
|   |-- mood/
|   |-- navigation/
|   |-- primitives/
|   `-- product/
|-- data/
|-- lib/
|-- pages/
|-- styles/
|-- App.tsx
`-- main.tsx

Core logic lives in src/lib:

• mood.ts: mood state, scoring, dates, summaries
• recommendations.ts: mood-state tool matching
• forecast.ts: transparent personal forecasting and pattern detection
• safety.ts: crisis-language detection
• storage.ts: local-first persistence
• export.ts: JSON export

Run Locally

npm install
npm run dev

Build:

npm run build

Tests:

npm test
npm run test:e2e

Legacy Python CLI tests:

python -m unittest discover -s tests -p "test_main.py"

Evidence-Informed Basis

Mood Compass is inspired by:

• Valence-arousal emotion models: https://pmc.ncbi.nlm.nih.gov/articles/PMC2367156/
• Positive and negative affect self-report concepts: https://pubmed.ncbi.nlm.nih.gov/3397865/
• Practical self-care guidance from NIMH: https://www.nimh.nih.gov/health/topics/caring-for-your-mental-health

Author

Mike Maeda
Computer Science & Data Analytics, Alfred University