Eugene Meidinger and Ruben Van de Voorde try things out:
Writing a skill takes time, and skills are context that can help but also hurt the quality of the agent’s work. So how do you know if a skill actually helps? You can test it with evals. An eval is a prompt that you re-run under the same conditions, so that the only variable is what you deliberately change. You use the same model, starting context, and the same files (and wipe leftover memory from previous attempts). Then you compare: before and after an edit, with and without a skill, or two versions of a skill. Agents are non-deterministic, so you have to run the comparison multiple times to get a reliable difference; a few runs per condition is enough to start. Decide how a run is scored before looking at the results: use objective checks where you can (did it run the right command?), and for subjective calls write down beforehand what a good answer must include. Also run in a clean workspace: if the solution is already in another branch, the baseline run isn’t a real test. Some companies run large automated eval suites, but a few saved prompts and some notes are enough to start.
Click through for a testing technique and a warning around relying on a generative model to do the work for you.