All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
2.0.0 - 2026-02-03
- Hybrid Search: Introduced hybrid retrieval strategies combining semantic and keyword (BM25) searches for improved response accuracy
- Feedback System: New
/api/feedbackendpoint to collect user ratings and comments on chat responses - Query Classifier: Intelligent query intent detection to optimize retrieval strategy
- Observability: Comprehensive logging for queries and feedback with analytics capabilities
- Model Selection: User-selectable AI models in chat form (gpt-4o-mini, gpt-4.1-mini)
- GitHub Webhook: Automatic documentation re-indexing via GitHub webhooks
- CORS Support: Cross-origin request handling with preflight support
- Streaming Response Buffer: Improved streaming with buffered line handling for data integrity
- Rate Limiting: Upstash Redis-based rate limiting for API protection
- Reranking: Search result reranking for improved relevance
- Reciprocal Rank Fusion: Result fusion from multiple search strategies
- Migrated from Hono to Next.js framework for improved functionality
- Enhanced Vercel configuration with Edge Runtime support
- Updated default chat model to gpt-4o-mini
- Improved README with comprehensive setup instructions
- Next.js 16 with App Router
- Vercel Edge Functions deployment
- Upstash Vector for embeddings storage
- Upstash Redis for rate limiting and caching
- OpenAI API integration for embeddings and chat