There is a quiet failure mode with AI agent skills that almost everyone hits: you build a custom skill, it looks perfect, and it simply never runs. No error, no warning. It just sits there unused. The reason is remarkably consistent, and so is the fix.
The failure is silent
Unlike a bug that throws an error, a skill that never triggers gives you nothing to debug. You ask for the thing the skill was built for, the agent does something reasonable without it, and you never realize the skill was skipped. That silence is what makes this so common — people do not even know it is happening.
The cause is almost always the description
An agent chooses which skills to use by reading their descriptions and matching them against your request. If the description does not match how you phrased things, the skill is skipped. And most descriptions are written too narrowly — as a label for the task rather than a map of how people ask for it.
