AI/ML
//Intermediate
Prompt Optimizer
Systematically improves prompts using structured outputs, few-shot examples, chain-of-thought placement, clear roles and measurable acceptance criteria, then A/B tests variants against a small eval set. Use when a prompt produces inconsistent or low-quality output, when onboarding a new model, or when the user wants to optimize, rewrite or debug a prompt with evidence instead of vibes.
Preview
# Prompt Optimizer Most prompts fail not because the model is weak but because the instructions are ambiguous. This skill improves prompts by tightening the contract — role, task, format, examples, and edge cases — and then proves the improvement with a befor
Related skills