fact-checking-explainers

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow is a fact-checking gate that reads “source passages” and “open[s] the actual code at the exact path and line” (SKILL.md:52-54, 68-69), which at runtime can include outsider-authored free text from externally gathered sources/pages or other non-user-authored documents, creating an indirect prompt-injection risk when those passages are ingested into the LLM context.