slide-decks

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly allows including public file_s3_keys and a "web_search_enabled" flag in the GenerateOutlineRequest and the workflows instruct the agent to generate an outline, poll job results, and then GET the deck/outline/slides (which may contain content pulled from public files or web search), so the agent will ingest untrusted third‑party content and act on it.