connect-ai-gmail

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted email content from Gmail, which presents an attack surface where malicious content in an email could attempt to influence the agent's behavior.\n
  • Ingestion points: The agent retrieves data from the Messages and Attachments tables as defined in SKILL.md.\n
  • Boundary markers: The skill provides a safeguard by advising the agent to obtain user confirmation before performing any action that modifies the mailbox (e.g., sending, replying, trashing, or relabeling mail).\n
  • Capability inventory: The skill utilizes stored procedures to send mail, update labels, and trash messages.\n
  • Sanitization: The instructions do not specify a requirement to sanitize or filter email body or subject text before interpretation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 02:10 PM
Security Audit — agent-trust-hub — connect-ai-gmail