review-bugs
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: 1.00). The prompt requires reading full files and including verbatim "evidence" code snippets and line-numbered excerpts in the report, so if those files contain API keys, tokens, or passwords the LLM would be forced to output those secret values verbatim.
Audit Metadata