Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already flag destructive, open-world, non-idempotent behavior, so the description's value is in the added context: availability of cancellation is decided externally and may fail, and the agent must never assert a delivered piece can be recalled. That honesty guardrail is genuinely useful beyond the annotation set, though failure modes and post-cancel state remain unstated.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.