Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint=true and readOnlyHint=false, but the description adds information annotations cannot express: retrying can resume real sends to real contacts and demands explicit approval. It does not explain what delete destroys (e.g. enrollments, logs) or whether unenroll is reversible, so a small gap remains.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.