⚠️ IMPORTANT DISCLAIMER This application is for TESTING AND EDUCATIONAL PURPOSES ONLY. DO NOT use this chatbot for medical advice or treatment of any medical condition. If you are experiencing a mental health crisis or medical emergency, please contact emergency services or a qualified healthcare provider immediately.

A conversational AI chatbot designed to explore thoughts and perspectives through structured dialogue, built with Claude AI.

This mental health chatbot provides:

• Multi-language support: English and German
• Personalized interaction: Adaptable communication styles (formal/informal)
• Structured conversation flow: Guided exploration of thoughts and beliefs
• Clean interface: Simple, distraction-free chat experience

• Language selection (English/German)
• Customizable communication style
• Session-based conversation tracking
• Built on Claude Sonnet 4.5 AI model

This chatbot is:

• NOT a replacement for professional mental health care
• NOT suitable for crisis intervention or emergency situations
• NOT intended for diagnosing or treating medical conditions
• FOR TESTING and educational exploration only

If you need professional help, please contact:

• Crisis hotline (US): 988 Suicide & Crisis Lifeline
• Emergency services: 911 (US) or your local emergency number
• Mental health professional: Therapist, counselor, or psychiatrist

• Python 3.8 or higher
• Anthropic API key (Get one here)

1. Clone the repository:

   git clone https://github.com/yourusername/mental-health-chatbot.git
   cd mental-health-chatbot

2. Install dependencies:

   pip install -r requirements.txt

3. Configure your API key:

   Create a .env file in the project root:

   cp .env.example .env

   Edit .env and add your Anthropic API key:

   ANTHROPIC_API_KEY=your_api_key_here
   

1. Start the app:

   streamlit run app.py

2. Access the app in your browser (usually at http://localhost:8501)

3. Select your language (English or German) from the sidebar

4. Start chatting - The chatbot will:

   • Ask how you'd like to be addressed
   • Adapt to your preferred formality level
   • Guide you through a structured conversation

• Language selection: Choose between English and German
• Communication style:
  • German: Du (informal) or Sie (formal)
  • English: First name (informal) or Title + Last name (formal)
• Session persistence: Conversation state maintained during your session
• Reset option: Start a fresh conversation anytime

mental-health-chatbot/
├── app.py                    # Main Streamlit application
├── core/
│   ├── orchestrator.py      # Main conversation orchestrator
│   ├── state_manager.py     # Session state management
│   ├── llm_interface.py     # Claude API interface
│   ├── method_handler.py    # Method-specific handlers
│   ├── rule_engine.py       # Rule-based dimension tracking
│   └── config_loader.py     # Configuration loader
├── config/
│   ├── config_en.py         # English language configuration
│   └── config_de.py         # German language configuration
├── requirements.txt          # Python dependencies
├── .env.example             # API key template
├── LICENSE                  # MIT License
└── README.md                # This file

Your API key should be stored in a .env file (see Setup above). Never commit your .env file to version control.

To modify the conversation experience:

• Language prompts: Edit config/config_en.py (English) or config/config_de.py (German)
• Conversation flow: Modify core/orchestrator.py and core/method_handler.py
• UI appearance: Customize app.py

• Orchestrator: Manages overall conversation flow and coordination
• State Manager: Handles session state and conversation history
• LLM Interface: Communicates with Claude API
• Method Handler: Processes method-specific conversation logic
• Rule Engine: Tracks conversation dimensions using rule-based approach
• Config Loader: Manages language-specific configurations

The chatbot guides users through several structured phases:

0. Introduction - Establish communication preferences
1. Problem Capture - Understand the topic to explore
2. Exploration - Ask clarifying questions
3. Identify Thoughts - Examine specific beliefs or perspectives
4. Behavioral Experiment - Design actions to test assumptions

• Model: Claude Sonnet 4.5 (Anthropic)
• Approach: Two-layer architecture
  • Layer 1: Python-based control logic and validation
  • Layer 2: LLM-powered conversational responses

Problem: API key not found or invalid

Solutions:

• Ensure .env file exists in project root
• Verify the file contains: ANTHROPIC_API_KEY=your_actual_key
• Check for typos in the environment variable name
• Verify your key at Anthropic Console
• Confirm you have available API credits

streamlit run app.py --server.port 8502

pip install --upgrade -r requirements.txt

1. Verify Python version: python --version (should be 3.8+)
2. Check all dependencies are installed: pip list
3. Review terminal error messages for specific issues

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

This project is licensed under the MIT License - see the LICENSE file for details.

• Built with Claude AI by Anthropic
• UI powered by Streamlit

This project is intended for educational and research purposes. When working with mental health applications:

• Always prioritize user safety and wellbeing
• Be transparent about AI limitations
• Never replace professional mental health care
• Implement appropriate safeguards and disclaimers
• Consider privacy and data protection carefully

Remember: This is a testing tool, not a medical device. Always seek professional help for mental health concerns.