Vague Triggers
Medium
- Confidence
- 82% confidence
- Finding
- The activation triggers are overly broad, using common phrases like “代码评审” and “CR” without constraining the task to a testing perspective. This can cause the skill to activate in unrelated review contexts and steer the agent into using its instructions and tools when the user intended a different workflow, increasing the chance of unintended code access or scope hijacking.
