external-llm-review

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly executes external LLM CLIs (e.g., codex exec and gemini --sandbox) and parses/structures their outputs as part of its required review workflow, thereby ingesting untrusted third-party model responses that can influence verdicts and next actions.