investigating-phishing-email-incident

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's code examples explicitly place API keys and bearer tokens into request headers/clients (placeholders like "YOUR_KEY" and "YOUR_VT_API_KEY" and f"Bearer {access_token}"), which requires inserting secret values verbatim and therefore creates an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). In scripts/agent.py, the runtime workflow ingests outsider-authored email text by loading a provided .eml file (parse_email_headers() and extract_urls_from_email() read the From/Subject/headers and decode the email body, then __main__ extracts URLs and calls URLScan/VirusTotal on them).