dbt Labs released a set of agent skills that enable AI coding agents to follow dbt best practices for analytics engineering. These skills help agents build models, troubleshoot issues, and understand complex workflows, making them more effective in data tasks. The skills are designed to evolve with community feedback and can be customized for specific organizational needs.

This article explains the differences between skills, commands, and rules in AI coding tools. It emphasizes how skills provide optional expertise for agents, while commands are explicit user instructions, and rules are fixed and always apply. The author also discusses effective strategies for organizing these elements to optimize performance.

The article discusses a study comparing two methods for teaching AI coding agents about Next.js: using skills and embedding documentation in an agents.md file. The results showed that the embedded documentation approach achieved a 100% pass rate, while the skill-based method struggled, highlighting the effectiveness of providing direct access to relevant information.

The article argues that development managers, who have focused on judgment and orchestration rather than coding, might be more valuable in a world where AI handles code production. As coding becomes nearly free, the emphasis shifts to understanding what to build and why, making managerial skills more relevant than technical ones. Managers who have honed their skills in specification writing, review processes, and business understanding are well-positioned for this new landscape.

The author discusses the evolution of their coding process with the introduction of a new plugin system called Superpowers for the Claude Code platform, which enhances the agent's capabilities through a skill-based system. This system allows for better organization and implementation of tasks, enabling Claude to self-improve and utilize skills effectively while also incorporating psychological principles of persuasion in its operation. The article emphasizes the importance of skills in empowering agents and the potential for further developments in this area.

The author expresses frustration with AI coding tools, feeling that despite their hype, they often lead to underwhelming results and time wastage compared to traditional coding methods. They struggle to reconcile their experiences with those of others who claim to achieve significant success using these tools, leading to a sense of inadequacy.

The article discusses the importance of designers learning to code in order to enhance their skills, improve collaboration with developers, and create more effective designs. It emphasizes the benefits of understanding the technical aspects of web development for designers in today's digital landscape.