Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Adds substantial context beyond the annotations: cost and payment mechanism ($0.01 per call, x402 USDC on Base, no account), a result cap (up to 200), matching semantics (exact full-name match, case- and diacritics-insensitive), and the scope limit that only *current* registrations are returned. Annotations cover safety/idempotency; the description covers everything else.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.