Show HN: Agent Skill + Activity Watcher = productivity hack for 2026 (github.com)

A new productivity enhancement tool integrates Claude Code with ActivityWatch to analyze weekly productivity using a sophisticated framework. This tool calculates focus scores, identifies repetitive app-switching patterns—termed "death loops"—and offers actionable insights tailored to improve work efficiency. Key features include smart auto-categorization of activities into productive, neutral, or distracting, and the ability to recognize effective AI-assisted coding workflows like Claude Code and GitHub Copilot. It provides dual scoring metrics for productivity and focus, facilitating a comprehensive understanding of work habits. This development is significant for the AI and machine learning community as it leverages local data analysis to deliver personalized productivity feedback without compromising user privacy. Users can customize activity categorization through JSON configurations, ensuring the tool adapts to individual workflows. With Python 3.8+ compatibility and easy integration via the optional aw-client API, this tool empowers users to evaluate their productivity patterns effectively and make targeted improvements, thereby enhancing their overall workflow efficiency. The ability to detect distractions and offer pragmatic solutions makes it a valuable resource for anyone looking to optimize their productivity in an increasingly digital workspace.