Vague Triggers
Medium
- Confidence
- 88% confidence
- Finding
- The description uses broad trigger phrases like 'reduce token usage,' 'audit context files,' 'optimize context window,' and 'slim down workspace files,' which can match many normal user requests and cause the skill to activate when not specifically intended. Unintended activation is risky here because the skill is designed to influence always-loaded context files, so an accidental invocation could lead to over-aggressive trimming or modification recommendations affecting persistent agent behavior.
