replicate-with-blocks

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

• Third-party content exposure detected (high risk: 0.85). The required runtime path ingests outsider-authored free text from the captured source HTML (e.g., html/*.html → liberate_section_extract(detail:"full") → section specs containing verbatim headings/body/buttons/reviews), which is then placed into the agent’s LLM context during reconstruction/building; this is indirect prompt-injection exposure from arbitrary third-party page content.