Teaches AI coding agents to create and maintain promptfoo eval suites. Encodes best practices like deterministic assertions first, file-based test organization, correct environment variable syntax, and proper hallucination/faithfulness checks. Includes a cheatsheet covering 30+ assertion types, provider patterns for OpenAI/Anthropic/Google/Bedrock/HTTP/Python/JS, and test patterns for JSON validation, dataset-driven scaling, and CI-friendly runs.