This document discusses an automatic text summarization system that aims to provide concise yet informative summaries of news articles and other texts to reduce information overload. It uses sentence extraction techniques like TextRank, which creates a graph of sentences based on word frequencies and relationships between sentences, to pick the most important sentences from an article to form the summary. The system is demonstrated on examples summarizing a beer running article, startup CEO article, and a New York Times article on a China-US air dispute, reducing the word counts significantly while conveying the main ideas. Areas for further development are also discussed, like improving the preprocessing and adding features to the graph-based ranking model.