Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds real behavioral context the annotations do not carry: which subsystems are checked (integrations, ingestion, clustering, gates, key scope), that the result is a playbook, that diagnostics cannot see the failed call, and that `problem` is filed verbatim in a report. However, it also discloses that share_report files a report — a state-changing side effect — while the annotations declare readOnlyHint=true, so an agent trusting the annotation alone could be misled.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.