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.
- Crawler support added.
- Performance improvements on data soruce indexing.
- Reduced container logs.
- GitHub codebase indexing improvements.
- Widget/Slack/Discord bot bug fixes.
- Guru Analytics page introduced.
- Settings page introduced.
- Reduce Dockerfile image size.
- Crawl4ai support added.
- Slack and Discord bots added.
- Fix backend init error by checking the github collection exists. Fixes #86.
0.1.1 - 2025-01-22
- Fix Binge graph. In some cases, the nodes are not visible.
0.1.0 - 2025-01-21
- Modern Next.js 14 frontend with TailwindCSS
- Django REST framework backend
- RAG system with advanced LLM techniques
- Multiple data source support:
- Website scraping with Firecrawl
- YouTube video transcription
- PDF document processing
- GitHub codebase indexing
- Vector similarity search with Milvus
- Message queue system with RabbitMQ for Celery
- Caching layer with Redis
- PostgreSQL database for data persistence
- Docker Compose based deployment
- Self-hosted installation script
- Binge feature for personalized learning paths
- Context evaluation system to minimize hallucination
- Comprehensive documentation:
- Installation guide
- Architecture documentation
- Development guidelines
- Website widget for embedding Q&A functionality
- Telemetry system with opt-out option
- Microservices architecture with Docker Compose
- Nginx for static file serving and reverse proxy
- Celery for asynchronous task processing
- Milvus for vector similarity search
- PostgreSQL for primary data storage
- Redis for caching and rate limiting
- RabbitMQ for message queue