code-simplifier

Fail

Audited by Snyk on Feb 15, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 0.80). The prompt mandates preserving exact functionality and autonomously refactoring recently modified code, so if that code contains hardcoded API keys, tokens, or passwords the agent would be required to reproduce those secret values verbatim in its output, creating an exfiltration risk.
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 08:59 PM