network-debugging

Fail

Audited by Snyk on Mar 4, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 1.00). The prompt explicitly shows and encourages embedding Authorization tokens in curl command-line headers (e.g., curl -H "Authorization: Bearer token"), which requires or encourages the agent to place secret values verbatim into generated commands/outputs, creating an exfiltration risk.
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 4, 2026, 05:19 PM