Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare a safe, idempotent, non-destructive read (readOnlyHint, idempotentHint, destructiveHint=false). The description adds genuinely non-obvious behavior: a free sample is returned with the paid URL and pay steps at no charge, while the full result costs $0.05 Base USDC via x402. That pricing/return-flow disclosure is above what the structured fields convey.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.