Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover destructiveHint=true, readOnlyHint=false, and idempotentHint=true. The description adds the effect on reports (revert to canonical value on regeneration) and the idempotent behavior on missing overrides, which is consistent with annotations and adds value beyond them.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.