regex-vs-llm-structured-text

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

• Third-party content exposure detected (high risk: 0.70). The skill’s runtime workflow ingests content: str (the “Source Text”) and then embeds that full outsider-authored text into the LLM prompt in validate_with_llm() via messages[0].content (“Text: {original_text}”), so if the operating user supplies any third-party/free-form document text, it becomes LLM-readable context (indirect prompt injection risk).