persona-customer-support
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it is designed to process untrusted external data from customer communications.
- Ingestion points: The skill triages and converts customer emails into tasks using the
gws gmailandgws workflowcommands. - Boundary markers: The instructions do not define specific delimiters or instructions for the agent to treat email content as untrusted.
- Capability inventory: The skill is granted access to read and write data across Google Workspace services (Gmail, Sheets, Chat, Calendar) via the
gwsCLI tool. - Sanitization: No explicit sanitization or validation of email body content is specified before the data is logged or escalated.
Audit Metadata