Natural-Language Policy Violations
Medium
- Confidence
- 89% confidence
- Finding
- The skill explicitly says it will 'apply their preferred language defaults,' which imposes a response style by default rather than only after a clear user request in the current interaction. This can reduce user autonomy and steer outputs in ways the user did not actively consent to, though in this skill the effect is limited to presentation style rather than secret exfiltration or privilege misuse.
