gws-workflow-meeting-prep

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data from Google Calendar events, such as agendas and descriptions. This creates a surface for indirect prompt injection where malicious content in a meeting invite could attempt to influence the agent's behavior. Ingestion points: Calendar event details (agenda, attendees, linked docs) referenced in SKILL.md. Boundary markers: None present. Capability inventory: Execution of the 'gws' CLI tool. Sanitization: None specified.
  • [COMMAND_EXECUTION]: The skill uses the 'gws' command-line tool, which is a vendor-owned resource for Google Workspace. While this involves executing a system command, it is part of the skill's primary intended purpose for productivity and is described as being used in a read-only context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 02:30 AM