agent-bom-vulnerability-intel

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow routes vulnerability lookups through agent-bom to public advisory databases (e.g., OSV/GHSA/NVD/EPSS/KEV) by sending package identifiers and CVE IDs; this constitutes outsider-authored free text/prose ingestion from public advisory sources when those lookups occur.