Auto-News is a personal productivity tool designed to aggregate and summarize information from diverse sources like RSS, Tweets, YouTube, web articles, and personal notes. It leverages Large Language Models (LLMs) such as ChatGPT, Gemini, and Ollama to distill content, filter noise, and generate actionable insights, catering to individuals aiming to stay informed and efficient in the AI era.

How It Works

The system aggregates content from various feeds and sources, processing them through an LLM backend. It can generate summaries, extract key insights, and even perform transcoding for YouTube videos lacking transcripts. The core advantage lies in its ability to significantly reduce information overload, providing users with concise recaps and actionable takeaways, thereby enhancing learning speed and productivity.

Quick Start & Requirements

• Managed Solution: Available via Web (Beta) and Mobile Apps (iOS/Android) in US/Canada.
• Self-Hosted: Requires Docker-compose, Portainer, Helm, or ArgoCD for installation.
• System Requirements (Self-Hosted):
  • OS: Linux or MacOS
  • CPU: Minimum 2 cores, Recommended 8 cores
  • Memory: Minimum 6GB, Recommended 16GB
  • Disk: Minimum 20GB, Recommended 100GB
• Documentation: https://github.com/finaldie/auto-news/wiki

Highlighted Details

• Supports multiple LLM backends: OpenAI ChatGPT, Google Gemini, Ollama.
• Features multi-agent capabilities for deep dives via web search and autogen.
• Aims to filter over 80% of content noise.
• Provides weekly top-k recaps and generates TODO lists from takeaways.

Maintenance & Community

• Users can open issues for questions or discussions.

Licensing & Compatibility

• The repository does not explicitly state a license in the provided README.

Limitations & Caveats

The mobile applications are currently limited to availability in the US and Canada. The web version is noted as being in Beta.