Vague Triggers
Medium
- Confidence
- 89% confidence
- Finding
- The trigger list includes broad, generic terms like "privacy" and "data privacy" that can match many unrelated conversations, causing the skill to activate outside its intended context. This creates prompt-surface expansion: users may receive unsolicited guidance or the model may route unrelated requests into this skill, increasing the chance of confusion, inappropriate advice, or interaction hijacking by an over-eager skill.
