A fully client-side, AI-powered fitness web application that generates personalised workout and nutrition plans based on a 12-step user assessment — with zero backend, zero frameworks, and zero dependencies.
FitForge takes a user through a 12-question assessment covering goals, body measurements, training experience, equipment, injuries, diet type, and lifestyle factors. It then produces a complete, tailored fitness blueprint including:
- Daily calorie and macronutrient targets (Mifflin-St Jeor + activity multipliers)
- A weekly workout schedule adapted to the user's training days and session duration
- A diet-specific meal plan (Standard / Vegetarian / Vegan)
- A BMI analysis and personalised health recommendations
- An achievement roadmap with milestone dates
The entire application is a single HTML page — no build tools, no bundler, no runtime dependencies.
- 12-Step Assessment — covers goals, body stats, activity level, training environment, injuries, diet, and lifestyle
- Science-Based Calorie Engine — Mifflin-St Jeor BMR with Harris-Benedict activity multipliers and goal-specific adjustments
- Protein-Anchored Macro Allocation — goal-driven protein targets (1.6–2.2 g/kg) with remaining calories split between carbs and fats
- Injury-Aware Workout Builder — automatically excludes flagged exercises based on declared injuries (knee, back, shoulder)
- Workout Split Generator — Full Body / Upper-Lower / Push-Pull-Legs splits based on available training days (1–7)
- Diet-Adapted Meal Plans — 18 goal × diet combinations (6 goals × 3 diet types)
- Exercise Library — 25 exercises with YouTube video tutorials, coaching cues, and muscle-group filtering
- Nutrition Library — 15 common foods with macronutrient breakdown bars
- BMI Calculator — standalone tool with animated gauge and category descriptions
- Progress Roadmap — milestone timeline with dates calculated from the current date
- Readiness Score — composite score based on sleep, stress, and hydration inputs
- Responsive Design — fully functional from 320 px to 4 K displays
- Accessible — semantic HTML5, ARIA roles, keyboard navigation, and screen-reader support throughout
- Print-Ready — dedicated print stylesheet hides navigation and isolates the results report
| Layer | Technology |
|---|---|
| Markup | HTML5 (semantic, ARIA-compliant) |
| Styling | CSS3 — custom properties, Grid, Flexbox, animations |
| Logic | Vanilla JavaScript (ES2020, no frameworks) |
| Fonts | Google Fonts — Space Grotesk + Inter |
| Images | Unsplash CDN (no local assets required) |
| Videos | YouTube embed (exercise tutorials) |
fitforge-gym/
├── index.html # Single-page application entry point
├── assets/
│ ├── css/
│ │ └── styles.css # Complete design system (~940 lines)
│ └── js/
│ └── app.js # All application logic (~1 450 lines)
├── .gitignore
├── LICENSE
└── README.md
No installation required. This is a static HTML project with no build step.
git clone https://github.com/ramiissa303/fitforge-gym.git
cd fitforge-gymOption 1 — Open directly in a browser
Double-click index.html
Option 2 — Serve locally (recommended for iframe embeds)
# Python 3
python -m http.server 8080
# Node.js (npx)
npx serve .Then visit http://localhost:8080 in your browser.
Add screenshots here after recording the live demo.
| Hero Section | 12-Step Planner | Results Dashboard |
|---|---|---|
![]() |
![]() |
![]() |
| Exercise Library | BMI Calculator | Nutrition Library |
|---|---|---|
![]() |
![]() |
![]() |
https://fitforge-planner.netlify.app
Hosted on Netlify — no build step, instant deploy.
https://github.com/ramiissa303/fitforge-gym
- Local storage persistence — save and resume the user's plan across sessions
- Dark mode — toggle with
prefers-color-schemedetection - Progressive Web App (PWA) — service worker + manifest for offline use and home-screen install
- PDF export — generate a shareable PDF of the results report
- Progress tracker — weekly weigh-in logging with a chart
- More exercises — expand the library to 100+ with full
mistakescoaching notes - Supplement recommendations — goal-aware supplement stack suggestions
- Multi-language support — i18n for Arabic, French, and Spanish
This project is licensed under the MIT License.
Rami Issa
- GitHub: @ramiissa303
- Email: ramiissa303@gmail.com
Built with vanilla HTML, CSS, and JavaScript — no frameworks, no dependencies.





