NewsBlur is a personal news reader that combines RSS feed aggregation with social features and machine learning-based content filtering.
The tool addresses the challenge of information overload by letting users train a classifier to automatically highlight or hide stories based on their preferences. It reconstructs full article text from truncated feeds, displays content in the original site's context, and enables real-time story delivery. The platform integrates social elements through shared story feeds and supports multiple consumption patterns including email newsletters and YouTube channels.
Self-hosting is the primary use case for this repository. The hosted service at newsblur.com offers free access up to 64 subscriptions with optional premium tiers, but developers who want complete data control or need to run a private instance can deploy using Docker and Docker Compose. The tool suits individuals and small teams seeking an alternative to commercial feed readers, particularly those valuing privacy and the ability to customize their news consumption environment. The technology stack—Django with Backbone.js frontend, PostgreSQL, MongoDB, Redis, and Elasticsearch—is mature and widely understood, making it accessible for self-hosted deployments.
Development activity shows consistent maintenance with regular updates addressing both core functionality and infrastructure. The project maintains native mobile applications across iOS and Android alongside the web interface, indicating sustained effort to support multiple platforms. Integration capabilities including IFTTT, third-party RSS clients, and an MCP server for AI agents demonstrate ongoing expansion of the tool's ecosystem. The codebase remains actively refined with attention to user-facing features like dark mode, multiple layout options, and search functionality.
