tad-generator

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow explicitly demands WebSearch/WebFetch availability and runs Phase 4 "Conduct 5 research rounds" (SKILL.md: "Confirm WebSearch and WebFetch tools available for research" and "Conduct 5 research rounds") and the tech-researcher outputs list external "references" (URLs), which shows the agent will fetch and interpret open/public web content that can influence architecture decisions and subsequent actions.