Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the annotations (readOnlyHint=false, idempotentHint=false), the description discloses real behavioral traits: an attempt cap of 3, returned feedback on failed attempts, zero reward for repeats after a pass, and the tiered reward structure (1–3 stars, 18 PP per star, badge at 2+ stars). These are exactly the side-effect and reward semantics an agent needs before invoking.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.