Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description discloses valuable dynamic behavior beyond annotations: returns cached canvas instantly if available, generates fresh analysis otherwise, charges one credit only for new content, and calls out non-read-only status. This gives the agent accurate expectations about side effects, costs, and performance.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.