behavioral-evals
Guidance for creating, running, fixing, and promoting behavioral evaluations. Use when verifying agent decision logic, debugging failures, debugging prompt steering, or adding workspace regression tests.
Install
git clone https://github.com/google-gemini/gemini-cli /tmp/gemini-cli && ln -s /tmp/gemini-cli/.gemini/skills/behavioral-evals ~/.claude/skills/behavioral-evalsFrom README
Behavioral Evals Overview Behavioral evaluations (evals) are tests that validate the agent's decision-making (e.g., tool choice) rather than pure functionality. They are critical for verifying prompt changes, debugging steerability, and preventing regressions. [!NOTE] Single Source of Truth: For core concepts, policies, running tests, and general best practices, always refer to evals/README.md. --- 🔄 Workflow Decision Tree Does a prompt/tool change need validation? No -> Normal integration tests. Yes -> Continue below. Is it UI/Interaction heavy? Yes -> Use appEvalTest (AppRig). See creating.md. No -> Use evalTest (TestRig). See creating.md. Is it a new test? Yes -> Set policy to USUALLYPASSES. No -> ALWAYSPASSES (locks in regression). Are you fixing a failure or promoting a test? Fixing -> See fixing.md. Promoting -> See promoting.md.