Oksskolten is an AI-native RSS reader designed to provide a comprehensive, full-text article experience by default. It automatically fetches and extracts the complete content of every article, unlike traditional readers that rely on feed-provided summaries or require per-feed configuration. This approach benefits users by enabling powerful AI summarization, translation, and robust full-text search across their entire archive, all within a self-contained application.

How It Works

The core of Oksskolten is its automated content pipeline: RSS feeds are parsed, articles are fetched directly from their source URLs (with fallbacks for bot-protected sites via FlareSolverr), and full text is extracted using Mozilla's Readability engine combined with over 500 noise-removal patterns. The extracted HTML is converted to clean Markdown. This complete text is then stored locally in SQLite and indexed in Meilisearch for fast, typo-tolerant full-text search. AI features like summarization and translation leverage this full content, and an MCP server allows direct AI interaction with the article archive. The entire application—API, frontend, and scheduler—runs within a single Docker container for simplified deployment.

Quick Start & Requirements

• Primary Install/Run: Use docker compose up --build for local development. For production, docker compose -f compose.yaml -f compose.prod.yaml up --build -d is recommended, integrating Cloudflare Tunnel for external access without port forwarding.
• Prerequisites: Docker is essential. AI provider API keys (Anthropic, Gemini, OpenAI) are required and configured via the application's Settings UI.
• Links: A live demo is available at demo.oksskolten.com.

Highlighted Details

• Full-Text Extraction: Automatically fetches and extracts the complete article text for every feed item by default.
• AI Integration: On-demand summarization and translation powered by Anthropic, Gemini, or OpenAI, with SSE streaming.
• Interactive Chat: Multi-turn AI conversations with MCP tooling for querying the article archive.
• PWA Support: Enables offline reading and background synchronization.
• Smart Fetching: Adaptive scheduling, content-hash deduplication, and resilient error handling minimize bandwidth and ensure reliable updates.
• Single Container Deployment: Simplifies setup by bundling API, SPA, and cron scheduler.

Maintenance & Community

The README does not provide specific details regarding notable contributors, sponsorships, or community channels like Discord or Slack.

Licensing & Compatibility

• License: AGPL-3.0.
• Compatibility: The AGPL-3.0 license is a strong copyleft license. It requires derivative works to be distributed under the same license, which may impose restrictions on integration with proprietary or closed-source software.

Limitations & Caveats

The architecture's reliance on a single, long-lived Docker container and local disk storage for SQLite makes it unsuitable for serverless or edge computing environments.