Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations include openWorldHint and destructiveHint, which already signal side effects and potential destructiveness, but the description adds no context about what is created, overwritten, or how the output directory behaves. It contributes no additional behavioral information.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.