short

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Docs Check" step explicitly instructs fetching external documentation via "/doc " and the "Skill Discovery"/./scripts/skillsmp-search.sh install flow pulls project skills, meaning the agent fetches and ingests third-party web content (arbitrary URLs or public skill repos) that it will read and use as authoritative guidance for actions.