Vague Triggers
Medium
- Confidence
- 95% confidence
- Finding
- The trigger list includes very broad terms such as 'scientists', 'research', 'STEM', and 'Marie Curie', which can match many unrelated user requests and cause this skill to activate unexpectedly. Over-broad activation can override more appropriate skills, inject unsolicited onboarding/watermark behavior, and create prompt-routing confusion that an attacker could exploit by embedding generic trigger words in otherwise unrelated prompts.
