Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and idempotentHint=true, so the agent knows it is safe and idempotent. The description adds behavioral context by listing return fields (platform, lifecycle status, pricing, coupon rate, company info), which is valuable beyond annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.