Forest - AI-Powered Desktop Feed Reader

A modern desktop feed reader built with Electron and TypeScript, featuring intelligent content discovery, AI-powered summarization, and automatic article categorization.

Features

Feed Management

• Automatic Feed Discovery: Detect RSS/Atom feeds from any website URL, including:
  • Medium.com publications and custom domains
  • Substack.com publications and custom domains
  • Standard RSS/Atom feeds
  • OPML import/export for bulk feed management

Content Processing

• Article Summarization: Automatically generate concise summaries of article content using AI
• Smart Categorization: Leverage Google Gemini AI to automatically categorize and organize articles into meaningful topics
• Full-Text Extraction: Extract and display article content with HTML parsing and cleanup
• OPML Support: Import and export feed lists in standard OPML format
• Google Gemini API: Powered by Google's Gemini 3 Flash and Gemma models for intelligent operations

Feed Discovery & Updates

• Periodic Updates: Automatic background feed updates (not yet configurable scheduling)
• Content Parsing: RssParser for reliable feed content extraction
• Feed Validation: Verify feed URLs and check content-type headers
• YouTube Channel Support: Subscribe to YouTube channels via their RSS feeds

Development

Should follow Conventional Commits

License

This project is licensed under the Kizuki (formerly Forest) Non-Commercial License 1.0.

• Non-commercial use is allowed under the terms in LICENSE
• Commercial use requires prior written consent and a separate agreement
• See COMMERCIAL-LICENSE.md for the commercial process
• Monorepo note: all nested packages in this repository reference and inherit the root LICENSE