reviewkit

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the reviewer to read repository files (including untracked source files) and to quote offending hunks in the report — and it names "secrets in code or logs" as a security finding — which would require the model to include secret values verbatim if they appear in the code.