StillPoint is an open-source mental health companion app designed to provide a private, personalized space for reflection and growth. It features an AI, powered by Google's Gemini models via Vertex AI, that learns and adapts to you over time through conversation and user-provided insights. Our goal is to create a supportive, non-judgmental partner for your self-discovery journey.

🌱 Why Open Source?

We believe that tools dealing with mental well-being demand the highest level of trust and transparency. StillPoint is open source because:

1. Transparency: Anyone can inspect the code to understand exactly how the AI works, how data is handled, and what prompts are used. No hidden algorithms or data usage.
2. Safety & Security: The community can audit the codebase for potential security vulnerabilities or privacy concerns, leading to a safer application for everyone.
3. Trust: Open development fosters trust. You can verify our commitment to privacy and ethical AI practices.
4. Community & Collaboration: We want to build StillPoint with the community. Open source allows others to contribute ideas, features, and improvements.

✨ Key Features

• Adaptive AI Companion: Engage in conversations with an AI that learns from your chat history, summaries, and questionnaire responses (securely stored in your private Firebase account).
• Personalized Experience: The AI tailors its interactions based on your preferences and past conversations, aiming for more relevant and insightful dialogue.
• Conversation Summaries ("Reflections"): Generate concise summaries of your chat sessions to capture key insights and track your journey over time.
• Voice Input: Speak your thoughts naturally. StillPoint uses OpenAI's Whisper API to transcribe your voice notes directly into the chat input.
• User Insights Questionnaire: Provide optional background information through a questionnaire to help the AI understand your preferences and tailor its approach from the start.
• Privacy-Focused: Built on Google Cloud (Firebase Authentication, Firestore, Vertex AI) with user data stored securely within individual user accounts. You own and control your data.
• Completely Open Source: The entire codebase is available for review and contribution.

🛠️ Tech Stack

• Frontend: React (with Vite)
• Styling: Material UI (MUI)
• Backend & Database: Firebase (Authentication, Firestore)
• AI Language Model: Google Gemini Pro (via Firebase Vertex AI SDK)
• AI Voice Transcription: OpenAI Whisper API
• State Management: React Hooks (useState, useEffect, useCallback, useMemo, custom hooks)
• Routing: React Router

🚀 Getting Started (for Developers)

1. Clone the repository:

   git clone https://github.com/AceCanacan/mental-health-app.git
   cd mental-health-app/app

2. Install dependencies:

   npm install
   # or
   yarn install

3. Set up Firebase:

   • Create a Firebase project at https://console.firebase.google.com/.
   • Enable Authentication (Email/Password recommended).
   • Enable Firestore Database.
   • Enable Vertex AI integration within your Google Cloud project linked to Firebase. Ensure the Gemini API (e.g., gemini-2.0-flash) is enabled.
   • Register a Web App in your Firebase project settings.
   • Copy your Firebase configuration credentials.

4. Set up OpenAI:

   • Get an API key from https://platform.openai.com/ for Whisper transcription.

5. Configure Environment Variables:

   • Create a .env file in the app directory (/mental-health-app/app/.env).
   • Add your Firebase and OpenAI credentials. Important: Vite requires environment variables exposed to the client to be prefixed with VITE_.

   # Firebase Configuration
   VITE_FIREBASE_API_KEY=YOUR_API_KEY
   VITE_FIREBASE_AUTH_DOMAIN=YOUR_AUTH_DOMAIN
   VITE_FIREBASE_PROJECT_ID=YOUR_PROJECT_ID
   VITE_FIREBASE_STORAGE_BUCKET=YOUR_STORAGE_BUCKET
   VITE_FIREBASE_MESSAGING_SENDER_ID=YOUR_MESSAGING_SENDER_ID
   VITE_FIREBASE_APP_ID=YOUR_APP_ID
   # VITE_FIREBASE_MEASUREMENT_ID=YOUR_MEASUREMENT_ID (Optional)
   
   # OpenAI API Key (for Whisper)
   VITE_OPENAI_API_KEY=YOUR_OPENAI_API_KEY
   
   # Google Cloud / Vertex AI (Gemini) - Note: Firebase SDK might handle auth implicitly if set up correctly,
   # but explicitly adding the key might be needed depending on setup/future changes.
   # VITE_GEMINI_API_KEY=YOUR_GOOGLE_CLOUD_API_KEY_WITH_VERTEX_AI_ACCESS (If needed)

   (Note: Ensure your Google Cloud project linked to Firebase has billing enabled for Vertex AI and OpenAI API usage.)

6. Run the development server:

   npm run dev
   # or
   yarn dev

   The app should now be running on http://localhost:5173 (or another port if 5173 is busy).

🤝 Contributing

Contributions are welcome! Whether it's bug reports, feature suggestions, or code contributions, please feel free to participate.

1. Issues: Check the Issues tab for existing bugs or feature requests. Feel free to open a new issue.
2. Pull Requests:
   • Fork the repository.
   • Create a new branch for your feature or bug fix (git checkout -b feature/your-feature-name or bugfix/issue-number).
   • Make your changes.
   • Ensure your code adheres to the project's style (ESLint is configured in eslint.config.js). Run npm run lint or yarn lint.
   • Commit your changes with clear messages.
   • Push your branch to your fork.
   • Submit a Pull Request to the main branch of the original repository.

📜 License

This project is licensed under the MIT License. See the LICENSE file for details. (You should add a LICENSE file with the MIT license text).

---

Thank you for your interest in StillPoint!