Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With readOnlyHint already present, the description adds substantial behavioral context beyond annotations: evidence sources, signed EIP-191 output valid 24 hours, special handling for ChainHelix agents as first_party with no verdict or signature, free unsigned access, and a 2-cent per-call price. It discloses cost, auth/signing requirements, and edge-case behavior clearly.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.