recipe-draft-email-from-doc

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection (Category 8).
  • Ingestion points: External content is retrieved via 'gws docs documents get' in SKILL.md.
  • Boundary markers: Absent; retrieved content is not delimited or accompanied by safety instructions to the agent.
  • Capability inventory: The skill has outbound communication capabilities via 'gws gmail +send' in SKILL.md.
  • Sanitization: Absent; document content is directly interpolated into the email body command without validation or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 01:28 PM
Security Audit — agent-trust-hub — recipe-draft-email-from-doc