gws-modelarmor

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the gws binary and uses it to execute shell commands for interacting with Google Model Armor APIs. It provides helpers for sanitization and template management.\n- [PROMPT_INJECTION]: The skill handles untrusted data by design as part of its filtering function, creating a surface for indirect injection attacks.\n
  • Ingestion points: User prompts and model responses provided to the +sanitize-prompt and +sanitize-response sub-skills.\n
  • Boundary markers: The SKILL.md does not specify explicit boundary markers or delimiters for the untrusted content it processes.\n
  • Capability inventory: The skill has the ability to execute shell commands via the gws CLI tool.\n
  • Sanitization: The primary purpose of the skill is to perform safety filtering using Google Model Armor templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 01:27 PM
Security Audit — agent-trust-hub — gws-modelarmor