Dev Tools · 1h ago
New CLI tool skilleval objectively tests Claude skill prompts
Developer Dileep Pandiya released skilleval, a CLI that evaluates Claude SKILL.md files via blind A/B testing. It runs tasks with and without the skill, using a randomized judge to eliminate position bias. The tool reports a margin-based score and warns when confidence is low due to insufficient runs.
Meridian48 take
A practical solution to a real problem, but its value depends on the quality of the tasks and judge model used.
claudeai-evaluation