gws-events

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill metadata specifies a requirement for the 'gws' binary, which is invoked to manage Workspace resources such as subscriptions, messages, and tasks.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes untrusted data from Google Workspace event streams. Ingestion points: Data retrieved via 'gws events' resource calls (messages, task updates). Boundary markers: None identified in the provided instructions. Capability inventory: The skill can create, delete, and modify subscriptions, as well as interact with active tasks and messages. Sanitization: No validation or sanitization of the event data is specified before it is returned to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 01:27 PM
Security Audit — agent-trust-hub — gws-events