cubrid-qa-fetch

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow uses the local cubrid-qahome-fetcher to fetch login-protected qahome pages (e.g., qahome.cubrid.org/qaresult/*.nhn) and saves the fetched HTML (raw/<page>.html) into the run directory, which is then used for LLM analysis—this is outsider-authored free text from a third-party web portal (qahome) not authored by the operating user.