Vague Triggers
Medium
- Confidence
- 89% confidence
- Finding
- The top-level description says the skill triggers when users ask to view schedules, add events, or remind upcoming events, but it does not clearly constrain when read-only versus state-changing actions are allowed. Broad activation language can cause an agent to invoke calendar access or reminder behavior in situations the user did not explicitly intend, increasing the risk of unauthorized reads or unintended actions.
