project-intelligence-change-event-extraction

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

• Third-party content exposure detected (high risk: 0.75). Step 1 instructs reading “every document in the Project Documents library” and Step 3/4 then extract and write their contents into chat/shareable output; if those documents include outsider-authored free text, that text is ingested by the agent’s LLM context for extraction and summarization.