Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already specify readOnlyHint, openWorldHint, idempotentHint, and destructiveHint, covering safety and idempotency. The description adds that it returns title, sheet/tab names, and properties, which is consistent but does not introduce new behavioral traits beyond what annotations imply.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.