Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond annotations (readOnlyHint, idempotentHint, destructiveHint=false), the description adds the date-range cap of at most 31 days, grouping by UTC day, and an explicit privacy/scope guarantee that it never returns member, trainer, service, attendance, or other participation details. These are meaningful behavioral disclosures not present in annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.