Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint and idempotentHint, so the description doesn't need to repeat those. It adds minimal behavioral context beyond stating the delete action and mentioning 'sensitive data,' but does not disclose additional traits like permissions or edge cases. The bar is lower with annotations, yet the added value is limited.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.