principle-never-block-on-the-human

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The prompt instructs agents to proceed without human approval and to make changes proactively (reserving confirmation only for a narrow class of "irreversible" actions), which encourages autonomous modifications to system state and could lead to privileged or state-changing operations even though it doesn't explicitly mention sudo or user-account changes.