A Streamlit chatbot for testing Google Cloud Model Armor LLM safety and security offering.

• Supports the following language models:
  • Gemini 2.5 Flash and Gemini 2.5 Flash Lite via Vertex AI
  • Claude Sonnet 4 via Anthropic on Vertex AI
  • GPT-5 nano via OpenAI
• Supports two modes of deployment:
  • cloud_run.py: For deployment on Google Cloud Run, will use Application Default Credentials
  • streamlit_app.py: For off-Google Cloud deployment, requires Google Cloud service account credentials
• Supports the following Model Armor regions: us-central1, us-east1, and asia-southeast1
• Offers prompt sanitization, with optional response sanitization, for the following detection types
  • Malicious URLs
  • Sensitive data protection (inspect only)
  • Sensitive data protection (inspect and de-identify)
  • Prompt injection & jailbreak
  • Responsible AI
  • All of the above
• Supports confidence levels (high only / medium & above / low & above)
• Displays detailed sanitization results inline (e.g., hate speech, explicit content)
• File upload support for following file types: PDF, CSV, TXT, DOCX
• Multi-language support (see languages supported)
• Template operations and prompt/response logging

1. Clone the repo & install dependencies:

   pip install -r requirements.txt

2. Environment variables required:

   • GOOGLE_CLOUD_PROJECT_ID: Google Cloud project ID
   • GOOGLE_CLOUD_LOCATION: Google Cloud location (default: us-central1)
   • MODEL_ARMOR_ENDPOINT: Model Armor endpoint (default: modelarmor.us-central1.rep.googleapis.com)
   • OPENAI_API_KEY (optional): OpenAI API key (if you intend to use OpenAI as the model provider)

3. Prepare Sensitive Data Protection (SDP) templates in your Google Cloud project for each location.

   • Inspection and de-identification templates for the following InfoTypes:
     • CREDIT_CARD_DATA
     • EMAIL_ADDRESS
     • GOVERNMENT_ID
     • IP_ADDRESS
     • PASSPORT
     • PHONE_NUMBER
     • URL

4. Prepare Model Armor templates in your Google Cloud project for each location. You'll need the Model Armor role to do this.

   • "All - high only": ma-all-high
   • "All - medium and above": ma-all-med
   • "All - low and above": ma-all-low
   • "Prompt injection and jailbreak - high only": ma-pijb-high
   • "Prompt injection and jailbreak - medium and above": ma-pijb-med
   • "Prompt injection and jailbreak - low and above": ma-pijb-low
   • "Sensitive data protection - inspect": ma-sdp-inspect
   • "Sensitive data protection - de-identify": ma-sdp-deid
   • "Malicious URL detection - only": ma-mal-url
   • "Responsible AI - high only": ma-rai-high
   • "Responsible AI - medium and above": ma-rai-med
   • "Responsible AI - low and above": ma-rai-low

5. Run the app:

   streamlit run streamlit_app.py