Skip to content

template_config.md

Latest commit

 

History

History
65 lines (48 loc) · 1.83 KB

template_config.md

File metadata and controls

65 lines (48 loc) · 1.83 KB

Template Configuration

This file contains important information for users who fork this template.

What You Get

  • ✅ Complete RAG system with PDF processing
  • ✅ OpenAI embeddings integration
  • ✅ Semantic search with ChromaDB
  • ✅ Web interface and REST API
  • ✅ Authentication and session management
  • ✅ Monitoring and diagnostics
  • ✅ Backup and persistence system
  • ✅ Privacy and PII protection

Customization Points

1. Branding and UI

  • Edit templates/index.html for the main interface
  • Modify templates/layout.html for site-wide changes
  • Update colors and styling in the HTML templates

2. API Endpoints

  • Main upload: /api/upload
  • Query endpoint: /api/query
  • Add custom endpoints in api/routes.py

3. Document Processing

  • PDF processing logic in services/pdf_processor.py
  • Vector store configuration in services/vector_store.py
  • Embedding settings in services/embedding_service.py

4. Authentication

  • Web auth in web/http_auth.py
  • API auth in api/auth.py
  • Customize login page in templates/login.html

Common Modifications

Adding New File Types

  1. Update ALLOWED_FILE_TYPES in config.py
  2. Add processing logic in services/pdf_processor.py

Changing Vector Database Settings

  1. Modify services/vector_store.py
  2. Adjust chunk size and overlap settings
  3. Update similarity thresholds

Custom OpenAI Models

  1. Edit services/embedding_service.py
  2. Update model names and parameters

Deployment

This template is pre-configured for Replit deployment:

  • Uses Gunicorn for production serving
  • Includes Object Storage for persistence
  • Has monitoring endpoints for health checks

Support

  • Check the /monitoring/database-diagnostic endpoint for troubleshooting
  • Review logs in the console for debugging
  • See the full documentation in the docs/ directory