Gws Modelarmor Sanitize Prompt

The gws-modelarmor-sanitize-prompt skill sanitizes user-generated prompts through Google Model Armor templates to detect and filter potentially unsafe content before it reaches AI models. It's designed for Google Workspace CLI users who need to implement inbound safety measures for user inputs, supporting both direct text input and stdin piping. The skill solves the problem of identifying and mitigating prompt injection attacks and harmful content in user submissions by leveraging pre-configured Model Armor templates.

skills.sh installs: 30.3K
GitHub stars: 32.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add googleworkspace/cli --skill gws-modelarmor-sanitize-prompt --agent claude-code

Installs into .claude/skills of the current project.

Files

modelarmor +sanitize-prompt

PREREQUISITE: Read ../gws-shared/SKILL.md for auth, global flags, and security rules. If missing, run gws generate-skills to create it.

Sanitize a user prompt through a Model Armor template

Usage

gws modelarmor +sanitize-prompt --template <NAME>

Flags

Flag	Required	Default	Description
`--template`	✓	—	Full template resource name (projects/PROJECT/locations/LOCATION/templates/TEMPLATE)
`--text`	—	—	Text content to sanitize
`--json`	—	—	Full JSON request body (overrides --text)

Examples

gws modelarmor +sanitize-prompt --template projects/P/locations/L/templates/T --text 'user input'
echo 'prompt' | gws modelarmor +sanitize-prompt --template ...