Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations provide readOnlyHint=false (implying mutation), but the description doesn't clarify what 'parse' entails or what the output looks like. The description adds the dual functionality of validation AND parsing beyond what annotations indicate, but doesn't explain behavioral aspects like error handling, format requirements, or what parsing produces.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.