gws-gmail-triage
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
gwscommand-line tool to interact with Gmail APIs. This is a legitimate vendor-provided tool (Google Workspace CLI) required for the skill's primary functionality. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and displays untrusted content from email subjects and senders.
- Ingestion points: Gmail message metadata (sender, subject) ingested via the
gwstool inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill markdown.
- Capability inventory: The skill is limited to reading and displaying email summaries; it does not have write or delete capabilities.
- Sanitization: No sanitization or filtering of email content is performed before presentation to the agent.
Audit Metadata