malware-hash

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

• Third-party content exposure detected (high risk: 0.70). This skill directly queries the public VirusTotal API at https://www.virustotal.com/api/v3/files/${QUERY_HASH} (response saved to /tmp/vt_response.json) and parses untrusted fields such as last_analysis_stats, last_analysis_results, tags, and meaningful_name to determine threat level and subsequent outputs, so third‑party content can materially influence decisions.