Nook is a web application designed to automatically collect and summarize tech news from various sources like Reddit, Hacker News, GitHub Trending, technical blogs, and academic papers. It targets tech professionals and researchers seeking to efficiently stay updated on the latest developments, offering a daily digest with interactive chat capabilities for follow-up questions, all at a low operational cost.

How It Works

Nook employs a modular, serverless architecture on AWS, utilizing Lambda functions triggered by EventBridge for scheduled data collection. It scrapes content from specified APIs (Reddit, GitHub, Hacker News, arXiv) and RSS feeds, then leverages Google Gemini API for Japanese summarization. Collected data is stored in S3, and a separate Lambda function provides a responsive web interface for browsing and interacting with the summarized content via a chat interface.

Quick Start & Requirements

• Installation: Clone the repository, set up environment variables (.env file with Gemini and Reddit API keys), install dependencies (pip install -r requirements.txt), and deploy using AWS CDK (make cdk-deploy).
• Prerequisites: AWS CLI with appropriate permissions, Python 3.11, Docker for building images, Google Gemini API key (paid), Reddit API keys (client ID, client secret, user agent).
• Setup: Deployment involves AWS infrastructure setup. Access the web interface via a URL provided after deployment.
• Links: GitHub Repository

Highlighted Details

• Supports diverse sources: Reddit, GitHub Trending, Hacker News, RSS feeds, arXiv papers.
• Interactive chat for follow-up questions on summarized content.
• Customizable information sources via TOML configuration files.
• Data stored in S3, enabling integration with other applications.

Maintenance & Community

• Project is actively maintained by discus0434.
• Contact via GitHub Issues for questions, bug reports, or feature requests.
• X (formerly Twitter): @IMG_5955

Licensing & Compatibility

• Licensed under GNU Affero General Public License v3.0.
• AGPLv3 is a strong copyleft license, requiring derivative works to also be open-sourced under the same license, which may have implications for commercial or closed-source integration.

Limitations & Caveats

The project relies heavily on AWS services, requiring significant AWS permissions for deployment. Operational costs are estimates and can vary based on usage. The AGPLv3 license may restrict integration into proprietary software.