This project provides a terminal-based news aggregation system that fetches and summarizes articles from various sources using multiple agents. It's designed for users interested in exploring agent-based workflows, terminal UIs, and leveraging tools like Amazon Q and tmux for parallel processing. The primary benefit is a self-contained, observable news summarization pipeline running entirely within the terminal.

How It Works

The system utilizes Amazon Q as an agent harness, with the Model Context Protocol (MCP) for parsing RSS feeds as tools. A main agent orchestrates the process by reading feed URLs, splitting them into chunks, and spawning three sub-agents in separate tmux panes. Each sub-agent processes a chunk of feeds, fetching content, parsing articles, and generating summaries, reporting back to the main agent upon completion. The main agent then consolidates these summaries into a final digest.

Quick Start & Requirements

• Install Amazon Q CLI: Follow the official guide.
• Set up AWS credentials.
• Clone repo and run:

  git clone https://github.com/eugeneyan/news-agents.git
  cd news-agents
  uv sync
  q chat --trust-all-tools
  context add --global context/agents.md
  

• Dependencies: Python, Amazon Q CLI, MCP.

Highlighted Details

• Demonstrates parallel agent execution within terminal using tmux.
• Leverages Amazon Q CLI for agent orchestration and MCP for tool integration.
• Generates individual and aggregated news summaries.
• Includes specific parsers for AI News, Hacker News, TechCrunch, Wired, and WSJ.

Maintenance & Community

The project is a personal experiment by eugeneyan. No specific community channels or roadmap are detailed in the README.

Licensing & Compatibility

The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

This project is described as a "little experiment" and may not be production-ready. Setup requires the Amazon Q CLI, which has its own installation and configuration steps. The specific licensing is not mentioned, which could impact commercial adoption.