Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds value beyond annotations by noting that human verification, review, and signature remain in the signing flow, indicating the tool does not complete the process. It also mentions the email matching condition. Annotations already indicate read-only and idempotent, so the description clarifies the logical guard.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.