Skills
skills/countbot-ai/countbot/email/Gen Agent Trust Hub

email

Fail

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: HIGHCREDENTIALS_UNSAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The configuration file scripts/config.json stores email authentication codes and passwords in plain text for QQ and 163 mail services.
  • [DATA_EXFILTRATION]: The send_email functionality in scripts/email_manager.py allows for reading arbitrary files from the filesystem and sending them as attachments, which could be used for data theft.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. Ingestion points: untrusted email content is retrieved via receive_emails and receive_emails_since in scripts/email_manager.py. Boundary markers: none are used to isolate email data from agent instructions. Capability inventory: includes deleting emails (delete_email) and sending emails with attachments (send_email). Sanitization: only basic HTML tag stripping is performed in _html_to_text, leaving the agent vulnerable to instructions embedded in the email text.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 11, 2026, 04:19 PM