Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations only indicate readOnlyHint=false, signaling a write operation, but the description adds no behavioral context such as what 'creating a file' entails, whether it overwrites existing files, what the token represents, or what side effects occur. It does not contradict the annotations, but it also does not enrich them.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.