Track. Improve. Repeat.
A modern, offline-first Progressive Web Application (PWA) for runners who mean business.
 Login |
Dashboard |
 Run logging |
Screenshot files to supply: docs/assets/login.png, docs/assets/dashboard.png, and docs/assets/run-logging.png.
RunRush is an offline-first running tracker for logging runs, following progress, and sharing milestones. Its Progressive Web App architecture supports local run logging when the network is unavailable and synchronization when connectivity returns.
- Activity tracking: Record distance, duration, pace, and estimated calories.
- Strava CSV import: Preview imported runs, detect duplicates, and process the batch transactionally.
- AI screenshot import: Upload a screenshot from another running app; the Gemini API extracts run details for review and import.
- AI run insights: Generate run feedback from activity details when the configured AI service is available.
- Weather integration: Add local weather details to runs when location is configured.
- Per-user ML prediction: Estimate a user's next run distance from their history.
GradientBoostingRegressoris used when there is enough data; smaller histories use a weighted-average fallback. - Training load (ACWR): Acute:Chronic Workload Ratio in the Analytics tab — compares your last 7 days of load to your 4-week average, with zone badges (detraining risk / sweet spot / caution / high spike) and an 8-week chart.
- Dashboard and analytics: View run history, charts, a calendar heatmap, goals, and progress indicators.
- Four-tab dashboard navigation: Home, Runs, Leaderboard, and Analytics are the four primary in-page tabs.
- Streaks, goals, and ranks: Track weekly goals and streaks, earn badges, and progress through distance-based ranks.
- Leaderboards and social feed: Follow runners, compare activity, and like runs.
- Installable PWA: Add RunRush to a supported mobile or desktop home screen.
- Offline run logging: Store queued runs in IndexedDB while offline.
- Background synchronization: Sync queued runs after connectivity returns, with duplicate checks.
- Service worker caching: Cache application assets and provide an offline fallback.
- Light and dark themes: Switch themes and persist a preference across sessions.
- Authentication and profiles: Use PIN authentication, profile settings, and Google sign-in when configured.
- Security protections: PINs are bcrypt-hashed; Flask-WTF CSRF protection covers state-changing requests with explicit route exemptions; Flask-Limiter rate-limits selected routes.
- Admin dashboard: Manage users and review platform activity.
RunRush uses a service worker and IndexedDB so users can queue a run while offline and synchronize it when the connection returns.
flowchart TD
A[RunRush PWA] --> B{Network Available?}
B -->|Yes| C[Flask Backend]
B -->|No| D[(IndexedDB)]
D --> E[Offline Run Queue]
E --> F{Connection Restored?}
F -->|Yes| G[Sync Engine JS]
G -->|Hash validation and duplicate check| C
C --> H[(PostgreSQL / SQLite)]
H --> I[Dashboard UI]
- Service worker: Caches application resources and handles offline fallback behavior.
- Offline queue: Stores runs in IndexedDB until they can be sent to the server.
- Data integrity: The sync endpoint checks payload hashes and duplicates before accepting queued runs.
flowchart LR
subgraph Client [Browser / PWA]
UI[Templates and JavaScript]
SW[Service Worker]
IDB[(IndexedDB)]
end
subgraph Server [Backend]
Flask[Flask Application]
Services[Application Services and AI]
ML[Run Prediction]
end
subgraph Data [Storage]
DB[(PostgreSQL or SQLite)]
end
UI <--> SW
SW -.-> IDB
SW <--> Flask
Flask <--> Services
Flask <--> ML
Flask <--> DB
| Layer | Technology |
|---|---|
| Backend | Python, Flask, Werkzeug, Jinja2 |
| Frontend | HTML5, CSS3, Vanilla JavaScript, Bootstrap |
| Database | PostgreSQL in production, SQLite for local development |
| Offline Storage | IndexedDB, Web Storage API |
| PWA | Web App Manifest, Service Worker Cache API |
| Data Visualization | Chart.js |
| ML | scikit-learn (GradientBoostingRegressor), pandas, NumPy |
| AI | Google Gemini API (google-genai) |
| Security | bcrypt, Flask-WTF/CSRF, Flask-Limiter |
| CI | GitHub Actions |
| Deployment | Gunicorn on Render |
RunRush/
├── .claude/launch.json
├── .env.example
├── .github/workflows/python-app.yml
├── .gitattributes
├── .gitignore
├── .python-version
├── LICENSE
├── pytest.ini
├── README.md
├── app.py
├── config.py
├── db.py
├── docs/
│ ├── archive/
│ │ ├── RUNRUSH_CODEBASE_AUDIT.md
│ │ ├── RunRush_Project_Report.html
│ │ ├── STEP4_INVESTIGATION.md
│ │ ├── database-compatibility-audit.md
│ │ ├── database-migration-map.md
│ │ └── improvements-history.md
│ ├── assets/
│ │ └── .gitkeep
│ ├── deployment.md
│ ├── developer-guide.md
│ └── project-summary.md
├── extensions.py
├── gunicorn.conf.py
├── marketing/
│ ├── runrush-qr-card.png
│ ├── runrush-qr.png
│ └── runrush-qr.svg
├── migrations/
│ ├── 001_initial_schema.sql
│ ├── 002_seed_badges_challenges.sql
│ ├── 003_add_user_onboarding_fields.sql
│ ├── 004_add_monthly_goals.sql
│ ├── 005_add_profile_emoji.sql
│ ├── 006_add_notifications.sql
│ ├── 007_add_monthly_summary_deliveries.sql
│ ├── 008_add_run_likes.sql
│ ├── 009_drop_user_pets.sql
│ ├── 011_add_monthly_report_views.sql
│ └── README.md
├── ml_predictor.py
├── models/
│ ├── __init__.py
│ └── user.py
├── requirements-dev.txt
├── requirements.txt
├── schema.sql
├── scripts/
│ ├── add_dummies.py
│ ├── apply_migration_006.py
│ ├── apply_migration_008.py
│ ├── apply_migration_009.py
│ ├── delete_user.py
│ ├── migrate_to_pg.py
│ └── smoke_script.py
├── services/
├── static/
│ ├── css/
│ ├── fonts/
│ ├── icons/
│ ├── img/
│ ├── js/
│ ├── manifest.json
│ └── sw.js
├── templates/
├── tests/
└── utils/
├── __init__.py
├── dates.py
├── decorators.py
├── rate_limiter.py
└── validators.py
- Android (Chrome): Open RunRush and choose the install prompt or select Install app from the browser menu.
- iOS (Safari): Tap Share, then Add to Home Screen.
- Desktop (Chrome/Edge): Use the install icon in the address bar when available.
RunRush uses SQLite for local development.
git clone https://github.com/Yajat-Sharma/Runrush.git
cd Runrush
python -m venv .venv
# Windows: .venv\Scripts\activate
# macOS/Linux:
source .venv/bin/activate
pip install -r requirements-dev.txtCopy .env.example to .env and set the required local values, including a SECRET_KEY. Then start the app:
python app.pyThe development server is available at http://localhost:5000.
See the Render deployment guide for service settings, database configuration, and the SQLite-to-PostgreSQL migration command.
- PINs are hashed with bcrypt.
- Flask-WTF CSRF protection is enabled, with explicit exemptions for selected endpoints.
- Flask-Limiter applies limits to selected sensitive routes.
- Backend validation checks submitted run data.
- Offline synchronization uses payload hashing and duplicate checks.
- Core run tracking and analytics
- Offline-first PWA architecture and background sync
- Four-tab dashboard navigation and light/dark theme system
- Global leaderboards, social feed, run likes, and distance-based ranks
- Strava CSV import and Gemini-powered screenshot import
- Per-user next-run distance prediction with a small-data fallback
- bcrypt PIN hashing, CSRF protection, and rate limiting
- GitHub Actions CI workflow
- Advanced training plans
- Direct OAuth integration with Strava/Garmin
- Granular heart-rate zone analysis
- Weekly team challenges
Contributions are welcome. Fork the repository, create a feature branch, commit your changes, and open a pull request.
Built with ⚡ by Yajat Sharma