Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Since annotations already declare readOnlyHint=true and destructiveHint=false, the description adds extra value by revealing the 'active' filter, the flat list structure, and how parent_id supports hierarchy reconstruction. This gives the agent a clear mental model of the response shape and filtering semantics, going beyond the annotated safety profile.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.