predict-issues

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs grep/read of matching file contents (e.g., "--output_mode content", "Grep exposed secrets/API keys", searching for process.env and API keys) and promises to show specific code locations, which would require the LLM to surface secret values verbatim if they appear in matches, so it mandates handling secrets in output.