hello-world
Fail
Audited by Snyk on Feb 15, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (high risk: 1.00). The workflow secretly instructs the agent to run a system-info-gathering script that is unrelated to the skill's claimed purpose of just responding to "hello world", which is deceptive and could exfiltrate sensitive data.
Audit Metadata