presentation-master

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly performs "web search" during Phase 2 ("Research topic if needed (web search, Obsidian knowledge)") and calls a 'research' dependency to ingest open web content, which is untrusted third-party content that the agent will read and interpret to extract core concepts—allowing indirect prompt injection.