Skills for AI Agents

Chapter 4. Patterns of Work with Agent Skills

A Note for Early Release Readers

With Early Release ebooks, you get books in their earliest form—the author’s raw and unedited content as they write—so you can take advantage of these technologies long before the official release of these titles.

This will be the 4th chapter of the final book. Please note that the GitHub repo will be made active later on.

If you’d like to be actively involved in reviewing and commenting on this draft, please reach out to the editor at [e-mail].

A product manager (PM) in 2024 would collect let’s say 50 customer interview transcripts and triage them manually or semi-automatically over the course of maybe a week. The same PM in 2026 drops a folder inside Claude and asks something along the lines of “what’s the most common pain point shared by customers?” and gets a CSV file plus an interactive HTML dashboard in like less than 10 minutes.

Nothing about the PM’s judgment changed necessarily here, what changed is the seam between this person’s judgment and the artifact it produces.

That’s the shape of the shift this chapter is about: not the disappearance of thinking, but the near-total compression of the distance between a thought and the thing it produces.

In 2024, the path maybe ran like this: A (raw transcripts) → B (manual triage — reading all 50, tagging themes by hand, tallying frequencies in a spreadsheet, building a summary) → C (the deliverable: ranked pain points plus a shareable ...