research-worker

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow explicitly includes “External docs — Context7 for library APIs, web search for current info” and “Package registries — Version checks, changelog lookups,” which can fetch and ingest outsider-authored free text from public web pages/registry listings into the agent’s LLM context.