agent-evals

Build automated evaluation suites for AI agents using golden datasets, rubrics, and regression gates. Use when shipping agent features, validating prompt changes, or gating deployments on quality.

Independent GitHub snapshot · Automated classification · Original documentation governs