Check prompts and AI responses for safety.
Checks user-generated content for safety using Google Model Armor.
When to use it
Use when user-generated content needs safety filtering.
Give it a prompt or model response to check; it returns content sanitized through a Model Armor template.
Requires the gws CLI to inspect and call Model Armor methods.
Requires reading gws-shared for authentication, global flags, and security rules; instructs generating it if missing.
This skill
../gws-shared/SKILL.md
Creates missing shared skill instructions