FitQuest Review
A React + TypeScript fitness frontend paired with an ASP.NET Core backend that generates personalized workout and nutrition plans via a multi-agent reasoning pipeline.
Each review covers deployability, value versus commercial alternatives, and privacy model. Tools that can run locally were started and exercised; mobile or backend-dependent tools were assessed from published builds, source code, and deploy guides. Ratings reflect what we were able to verify.
FitQuest — Multi-Agent AI Fitness Coaching Platform
A React + TypeScript fitness frontend paired with an ASP.NET Core (.NET 10) backend that generates personalized workout and nutrition plans via a multi-agent reasoning pipeline. Last updated: 2026-06-21.
One-sentence verdict: A polished hackathon-style fitness coach with a visible AI reasoning chain; the frontend runs easily, but the backend requires .NET 10 and an AI key to unlock full functionality.
What the System Is
FitQuest is a full-stack fitness coaching demo split across two repositories: a React 18 + TypeScript frontend (fitquest-frontend) and an ASP.NET Core backend (fitquest-backend). The frontend exposes the AI's reasoning process step-by-step, showing how agents analyze recovery score, sleep, training history, and daily check-ins before generating a plan. The backend implements two agents — a training plan agent and a nutrition agent — using an OpenAI-compatible API (Azure AI Foundry or DeepSeek).
The deploy guide reports the frontend was started on port 5173. It also notes that the backend was not configured with an AI key in this environment, so the screenshots show the static frontend UI and some backend-dependent features display error messages.
| Key data | |
|---|---|
| Category | Fitness & Movement |
| Language | TypeScript / React + C# / ASP.NET Core |
| License | MIT (frontend and backend) |
| Self-hosted | Yes |
| AI provider | Azure AI Foundry or DeepSeek (OpenAI-compatible) |
| Database | SQLite (dev) / PostgreSQL or Supabase (production) |
| Auth | JWT Bearer tokens |
How to Install and Deploy
Backend (optional; frontend shows errors without it)
cd /data2/docker/going-global/repos/fitquest-backend
cp .env.example .env
# Edit .env: DATABASE_PROVIDER=sqlite, AI_API_KEY, AI_BASE_URL, AI_MODEL
dotnet run
# → http://localhost:3001
Frontend
cd /data2/docker/going-global/repos/fitquest-frontend
npm install
npm run dev
# → http://localhost:5173
There is no top-level Docker Compose that launches both services together.
How to Test
The documented test flow is:
- Open
http://localhost:5173. - Register or log in (requires a running backend).
- Submit a daily check-in (sleep, energy, stress, weight) and confirm the recovery score appears.
- Generate an AI workout plan and watch the reasoning chain animate.
- View the weekly schedule, training history, and nutrition page.
If the backend is missing or lacks an AI key, plan generation and nutrition advice will show error banners.
Privacy & Compliance
FitQuest is not HIPAA compliant. The backend stores user profiles, check-ins, and training sessions in SQLite by default. AI requests are sent to the configured provider (Azure or DeepSeek). LocalStorage is used on the frontend for caching. Do not enter health data you are not comfortable sending to those providers.
FitQuest vs Commercial Fitness Coaches
| Dimension | FitQuest | Commercial Apps (e.g., Freeletics, Nike Training Club, Future) |
|---|---|---|
| Cost | Free / self-hosted + API usage | Freemium or subscription |
| AI reasoning visibility | Live animated reasoning chain | Black-box recommendations |
| Setup effort | Medium; needs Node.js + .NET 10 + AI key | Mobile app signup |
| Workout library | AI-generated; no fixed exercise database | Large curated libraries |
| Wearable sync | None | Apple Health, Garmin, Fitbit, etc. |
| Social / community | None | Often included |
| Customization | Full source code | Closed |
| Offline support | None; backend-dependent | Varies |
Who Should Use It
- Developers interested in a multi-agent reasoning UI for fitness.
- Teams evaluating how to expose AI decision steps to end users.
- Hackathon participants looking for a full-stack React + .NET fitness scaffold.
Who Shouldn't Use It
- Users wanting a plug-and-play daily fitness app.
- People who need wearable integration, social features, or structured programs.
- Organizations needing HIPAA-compliant or clinically validated coaching.
FAQ
Is FitQuest free?
Yes. Both repositories are MIT-licensed. You pay for your own server and any Azure AI Foundry or DeepSeek API usage.
Do I need the backend?
The frontend can run without the backend, but login, plan generation, nutrition advice, and session saving require it.
What AI models does it use?
The backend supports Azure AI Foundry or DeepSeek via an OpenAI-compatible endpoint. The model is configured through environment variables.
Verdict
FitQuest is a well-presented fitness-coaching prototype with a distinctive reasoning-chain UI. The frontend is easy to run, but the full experience depends on a .NET 10 backend with a valid AI provider key. It is a useful learning project and demo, not yet a daily-use replacement for commercial fitness apps.
Ratings: Deployability 3/5 · Value vs Commercial 3/5 · Privacy Compliance 3/5