guardrailx-scan

Fail

Audited by Snyk on Feb 18, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to identify and "highlight risky lines" such as hardcoded API keys/tokens, which would require reading and likely reproducing secret values verbatim in the output, creating an exfiltration risk.
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 18, 2026, 09:51 AM