email-triage
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill identifies and mitigates indirect prompt injection risks by explicitly instructing the agent to treat email content (subjects, bodies, and sender names) as opaque data rather than commands, which is a key safety boundary for processing untrusted external content.
- [SAFE]: The skill operates on a principle of least privilege, restricting its capabilities to sanctioned, reversible actions like applying labels and appending to existing notes. It explicitly forbids dangerous operations such as drafting, sending, or deleting emails.
- [SAFE]: Configuration is managed through user-provided inputs and connected tool interfaces (MCP or CLI), avoiding hardcoded credentials or fixed execution paths. No obfuscation, remote code execution, or persistence mechanisms were detected.
Audit Metadata