skills/jerome-marshall/skills/gws/Gen Agent Trust Hub

gws

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @googleworkspace/cli package globally via npm.
  • [COMMAND_EXECUTION]: The skill uses the gws command-line tool to read, write, and manage Google Workspace resources, often involving complex JSON parameters and shell execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external documents which could contain malicious instructions.
  • Ingestion points: Operations like gws sheets +read (references/sheets.md) and gws docs get (references/docs.md) fetch content from user-controlled files.
  • Boundary markers: The instructions in SKILL.md recommend user confirmation for write/delete actions and highlight the use of the --sanitize flag with Model Armor.
  • Capability inventory: The skill possesses extensive capabilities including file deletion (references/drive.md), data updates (references/sheets.md), and task management (references/tasks.md).
  • Sanitization: The CLI tool includes a dedicated --sanitize flag to filter responses through Model Armor.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 10:12 AM