research

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly uses WebSearch/WebFetch (and a codex URL fallback) to fetch and validate content from public websites and community sources (e.g., GitHub issues, Stack Overflow), so the agent will ingest and interpret untrusted, user-generated third-party content as part of its workflow.