Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds detailed behavioral rules beyond the annotations: lowercase, accent stripping, collapsing to a single hyphen, trimming edges. This is exactly the kind of context an agent needs to predict output. No contradiction with readOnlyHint/idempotentHint.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.