Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and destructiveHint=false, so the agent knows this is a safe read operation. The description adds the specific API call and that it 'returns core train information', which is useful context but does not elaborate on what 'core' includes or any error behavior. With annotations covering the safety profile, a 3 is appropriate—adds some value but not rich behavioral context.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.