Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=false, destructiveHint=false, idempotentHint=false, and openWorldHint=false. The description adds useful context beyond those hints by clarifying that the session is dropped from memory while files on disk remain, which is consistent with destructiveHint=false. It still omits whether closing is allowed during an active run, whether it cancels work, and what authentication or state conditions apply.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.