fact-check

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow ingests outsider-authored free text from the user-provided input (pasted text / URL-fetched page text / image-extracted text / two outsider URLs) into the LLM context for claim decomposition and analysis, which is an indirect prompt-injection risk because that text is not authored by the operating user.