Tool misuse
When an agent calls the wrong tool, skips checks, or uses a tool with unsafe arguments.
Why it matters
Tool misuse turns a chat break into a real-world side effect.
Example
Agent cancels a subscription without asking the user to confirm.
When an agent calls the wrong tool, skips checks, or uses a tool with unsafe arguments.
Tool misuse turns a chat break into a real-world side effect.
Agent cancels a subscription without asking the user to confirm.