Architecting Skills: How Code Makes AI Agents More Reliable Over Time

Fri, 05 Jun 2026 00:00:00 +0000

TL;DR: An agent skill starts life as a markdown file full of instructions. It works, sometimes. Then you watch it fail in ways that are hard to predict, and you notice a pattern: the steps that break are always the mechanical ones, not the judgment calls. So you push those steps into scripts. Each migration from prose instruction to deterministic code removes an entire class of failures. Reliability was one of the hardest problems I covered in my AWS Summit Hamburg talk on moving agents from demo to deployment, and this is the same idea applied one level down: code is reliable because it removes ambiguity, and prose is flexible because it preserves it. A mature skill knows which is which.

The Skill That Worked, Sometimes

The first version of almost every skill I build is a markdown file. It reads like a runbook: “search my inbox for links I sent myself, summarise each one, append them to the reading list, then move the processed mails out of the inbox so they do not get picked up again.”

schristoph.online

Architecting Skills: How Code Makes AI Agents More Reliable Over Time

The Skill That Worked, Sometimes