Thysrael/Horizon
Horizon
Horizon is an AI-powered news aggregation system designed to build custom pipelines that fetch, filter, and enrich information from diverse web sources. It utilizes large language models to automate information filtering, scoring content to remove noise and highlight high-value stories.
The system integrates the Model Context Protocol to expose pipeline stages as tools for external AI assistants. It employs a unified adapter to standardize diverse AI model providers for consistent content scoring and summarization tasks.
The pipeline aggregates data from RSS feeds, social platforms, financial toolkits, and code repositories. It manages content through deduplication, quota-based category filtering, and contextual enrichment before delivering multilingual briefings via email, webhooks, or static site deployment.
Workflows are orchestrated through recurring cloud automation to manage the scheduled collection and delivery of processed information.
Features
- Multi-Source Content Aggregation - Aggregates technical content from diverse sources like RSS, social platforms, and repositories into a unified structure.
- News Aggregators - Collects and categorizes content from social media, RSS, and financial feeds into a centralized AI-powered stream.
- Model Provider Integrations - Provides a unified interface for connecting and configuring multiple large language model providers for scoring and summarization.
- AI Content Scorers - Uses AI models to score content and exclude items that fall below a specific importance threshold.
- AI-Powered Content Filters - Uses AI-driven systems to filter raw web content and isolate high-value stories from noise.
- LLM Provider Adapters - Standardizes diverse AI model provider interfaces into a single internal adapter for consistent scoring.
- Content Aggregators - Collects and consolidates news from code repositories, community forums, and RSS feeds into a single stream.
- Multilingual Briefing Generators - Creates summarized daily reports from aggregated data in multiple languages for diverse audiences.
- Semantic Content Deduplication - Identifies and merges identical stories across multiple platforms using semantic content deduplication.
- Content Processing Pipelines - Builds custom workflows to fetch, deduplicate, and enrich data from diverse web sources before final delivery.
- AI Information Filters - Employs large language models to score and filter news content, removing noise to highlight high-value stories.
- Workflow Tool Exposures - Exposes pipeline workflow definitions as callable tools that AI agents can discover and invoke.
- Model Context Protocol Integrations - Exposes pipeline stages as tools for external AI assistants using the Model Context Protocol.
- Model Context Protocol Servers - Implements the Model Context Protocol to expose pipeline stages as tools for external AI assistants.
- Recurring Task Automation - Schedules data aggregation and delivery processes as recurring tasks via cloud automation services.
- Source Filters - Provides a configuration system to isolate content from specific publishers or feed sources based on defined criteria.
- Content Metadata Enrichment - Augments stories with web-researched background information and community comment summaries for deeper context.
- Story Deduplicators - Merges identical stories appearing across different platforms to ensure each piece of news appears only once.
- Financial News Searchers - Retrieves equity and macro news from multiple providers through a standardized financial data toolkit.
- Quota-Based Limits - Balances topic variety by applying individual limits to specific content groups during aggregation.
- Cloud Orchestration & Automation - Utilizes cloud orchestration and automation to trigger recurring content fetching and delivery tasks.
- Email Summary Deliveries - Sends daily briefings via mail servers and manages user subscriptions by scanning incoming emails.
- AI Tool Definition Servers - Implements a server that communicates tool capabilities to AI assistants for managing content workflows.
- Multi-Channel Notification Dispatch - Dispatches generated reports through multiple channels including email, chat platforms, and webhooks.
- AI-Enhanced Aggregation - Personal AI news radar for daily briefings.
Star history
How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.
