Vague Triggers
Medium
- Confidence
- 90% confidence
- Finding
- The trigger text says to use this skill for ANY Benchmark Email request and instead of calling the API directly, which is overly broad and can cause the agent to route unrelated or higher-risk Benchmark Email tasks through this skill without sufficient task-level scrutiny. Broad auto-invocation language increases the chance of inappropriate tool use, especially if future actions are added or if a user request mixes read and write intents.
