docs/changelog-doc.md

Changelog

All notable changes to Anveshak: Spirituality Q&A will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[1.0.0] - 2025-04-01

Added

• Initial release of Anveshak: Spirituality Q&A
• Core RAG functionality with E5-large-v2 embedding model
• FAISS index for efficient text retrieval
• Integration with OpenAI API for answer generation
• Streamlit-based user interface
• Caching mechanisms for improved performance
• Support for customizable number of sources and word limits
• Pre-selected common spiritual questions
• Comprehensive acknowledgment of sources and publishers
• Detailed documentation

Technical Features

• Google Cloud Storage integration for data storage
• Authentication handling for GCP and OpenAI
• Memory optimization for resource-constrained environments
• Multi-page Streamlit application structure
• Custom CSS styling for enhanced user experience
• Privacy protection with no user data storage
• Concise answer generation system
• Recognition of Saints and Spiritual Masters of all backgrounds and traditions

Future Roadmap

Planned for v1.1.0

• Multi-language support (Sanskrit, Hindi, Bengali, Tamil, and more)
• User feedback collection for answer quality
• Enhanced answer relevance with hybrid retrieval methods
• Additional spiritual texts from diverse traditions
• Improved citation formatting with page numbers where available

Planned for v1.2.0

• Self-hosted open-source LLM integration
• Advanced visualization of concept relationships
• Search functionality for specific texts or authors
• Audio output for visually impaired users
• Mobile-optimized interface

Planned for v2.0.0

• Meditation timer and guide integration
• Personalized learning paths based on user interests (implemented with privacy-preserving approaches like client-side storage, session-based preferences, or explicit opt-in)
• Interactive glossary of spiritual terms
• Spiritual practice guide with scheduler and tracker
• Community features for discussion and shared learning