Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond annotations that already flag destruction, open-world behavior, and non-idempotency, the description discloses the invitation/acceptance workflow, return fields, unchanged billing for the sender, the keep_editor default, and workspace eligibility constraints. These are meaningful behavioral details an agent would not get from annotations alone.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.