Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, and destructiveHint, so there is no safety contradiction. The description adds the crucial behavioral fact that the tool returns a user-facing authorization URL rather than direct file contents, which is essential for setting agent and user expectations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.