gws-workflow-meeting-prep

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a local command-line tool, gws, to fetch meeting agendas, attendee lists, and document links. This execution is scoped to its primary function of meeting preparation.
  • [SAFE]: The skill is explicitly documented as read-only, which mitigates the risk of unintended data modification or deletion.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from external sources, specifically meeting descriptions and attendee names.
  • Ingestion points: Meeting agendas, descriptions, and linked documents processed by the gws workflow in SKILL.md.
  • Boundary markers: No explicit delimiter or instruction-guarding markers are present to isolate the fetched content from the agent's core instructions.
  • Capability inventory: The agent utilizes the gws CLI to query calendar data.
  • Sanitization: There is no evidence of sanitization or filtering applied to the meeting data before it is presented to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 01:27 PM
Security Audit — agent-trust-hub — gws-workflow-meeting-prep