Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnly, non-destructive, not open-world), so the bar is lower. The description adds genuine context beyond them: the tool is state-dependent on previously retrieved results, performs no search, and is intended for a separate rendering step in a client pipeline. It says nothing about what the rendering produces, but the output schema exists to carry that.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.