brand-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill accepts arbitrary pasted files/URLs as input and explicitly requires showing "before" (original) text in revised sections, which would force the LLM to echo any secret values present in the input verbatim, creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow accepts “Content to review” from pasted text, a file path, or a URL, which at runtime can cause the agent to ingest outsider-authored free-form text (e.g., arbitrary web page content or downloaded/pasted third-party material) into the LLM context for review.