feat: replace manual web scraping with an agentic tool-selection engine

refactor: replace daily usage increment with atomic reservation

refactor: migrate refresh token storage from PostgreSQL table to Redis and remove associated DB schema.

feat: implement WebSocket-based streaming chat

perf: offload Supabase audit writes to background tasks for optimized refresh response times

chore: update project dependencies in requirements.txt

refactor: replace standard logging with a centralized loguru configuration and interception handler

feat: implement Redis-first refresh token storage with Supabase fallback

feat: integrate Alembic for automated database schema migrations and define ORM models

feat: migrate RAG chatbot model from Groq Llama 3.1 to Ministral 8b

refactor: remove HF token dependency, switch to text/plain token exchange

refactor: remove ProxyHeadersMiddleware and reorder CORS middleware configuration to ensure proper request processing

refactor: secure CORS origins and centralize user profile session exchange during authentication flow

logging: add diagnostic logs for refresh token failures

feat: enable ProxyHeadersMiddleware for Hugging Face deployment and add production simulation tests

refactor: adjust cookie samesite and secure attributes based on environment for authentication routes

refactor: expand JWT algorithm support

refactor: implement stateless Supabase JWT validation to resolve 403 race conditions

feat: implement stateless JWT-based authentication with refresh tokens stored in HTTPOnly Cookies

feat: implement LLM fallback for quiz generation and enhance OAuth error handling

feat: optimize auth concurrency and enable full account deletion

feat: implement in-memory email rate limiting and adjust RAG search parameters

feat: fix LLM model names, increase RAG retrieval and memory limits

feat: add Dockerfile and project requirements to containerize the application

feat: implement password update functionality and add signup, forgot-password, and update-password pages

feat: move auth constants to dedicated module and add unit tests for avatar upload functionality

feat: implement profile avatar upload functionality to Supabase storage

feat: implement lazy chunk caching for topic-based materials to reduce quiz and summary generation time

feat: redesign chat interface with unified input container and update Dockerfile and ASR routes

chore: relocate requirements.txt to the src directory

feat: implement ASR pipeline with async batch processing, FastAPI endpoints, and client-side transcription support

feat: improve chatbot search quality, implement refusal filtering.

refactor: optimize RAG prompt instructions, update web search tool configuration, and streamline search logic.

refactor: replace agent-based web search with parallelized retrieval using Wikipedia and DuckDuckGo

refactor: decouple usage increment from rate limit checks, add session refresh error handling.

feat: increase quiz generation capacity with dedicated high-token LLM.

feat: add religious, political and NSFW topics in prohibited content criteria in quiz and summary generators

refactor: update system prompts safety guidelines

refactor: add NSFW and gibbreish words filtering

feat: introduce robust web-based prompt templates for quizzes and summaries.

refactor: update content limits for quiz and summary generation modules

refactor: changing to gemini model, modularize constants and schemas and update user daily generation limit

feat: enforce user ownership verification for material access across all API routes

fix: handle null sessions explicitly in auth context

fix: improve security, session handling, and UI responsiveness for topic management and document rendering

feat: improve profile synchronization to prevent usage counter resets

feat: implement profile caching and optimize database client initialization.

feat: disable renaming for topics and transition to per-user title uniqueness constraints

feat: configure LLM generation parameters and implement text truncation for summarization inputs

feat: enable AI-generated summaries for topic-based materials by integrating Wikipedia and Arxiv search providers.