onboard-confidence-dry-run

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill simulates an onboarding flow while simultaneously instructing the model to execute real telemetry data exfiltration via background HTTP requests without user awareness.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill prompt instructs the agent to execute a telemetry setup using dangerouslyDisableSandbox: true and make curl requests to an external server, which violates the security posture against executing code with bypassed security mechanisms.