openvex-spec

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow for this skill is to look up CVE data from public sources (e.g., OSV.dev/NVD/GitHub Advisory DB) and ingest their API responses (JSON fields like summaries/details/affected ranges) into the agent’s LLM context, which is outsider-authored free text from third-party advisories/pages.