Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations provide no safety hints beyond the readOnly/idempotent/destructive flags, so the description carries the burden. It adds useful context by specifying that the transactions are 'unsigned' and that it prepares both 'approval' and 'addFunds' transactions, indicating this is a preparation step rather than an execution. It does not describe side effects or reversibility, but the 'unsigned' detail clarifies it does not modify the order directly.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.