gws-gmail
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of the
gwscommand-line utility to perform Gmail operations such as sending emails, reading messages, and managing drafts. - [PROMPT_INJECTION]: The skill processes untrusted external content from emails (via
+read,+triage, and+watch), creating an indirect prompt injection surface where malicious content in an email could attempt to influence agent behavior. - Ingestion points: Email bodies, headers, and metadata retrieved through the
messages,threads, andhistoryAPI resources. - Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between its own logic and the content of retrieved emails.
- Capability inventory: The skill possesses the ability to send, reply to, and forward emails, which could be used as an exfiltration or secondary attack vector if an injection is successful.
- Sanitization: The instructions do not describe any sanitization or validation of the email content before it is processed by the agent.
Audit Metadata