gws-modelarmor
Helps your agent filter user-generated content for safety. Protects against inappropriate or harmful content.
What it does
This command helps keep AI-generated content safe and appropriate. It acts as a filter to ensure AI responses do not contain dangerous or unsuitable information.
When to use it
- ✓When you want to be sure an AI's response is safe.
- ✓When creating content that will be seen by many people.
- ✓To prevent the AI from generating something offensive or harmful.
Example of what to ask your AI
“Check if this AI response is safe to publish.”
Want more? Open the sections below.
How to use
The easiest way: copy the message below and paste it into Claude, ChatGPT, Cursor or any AI. It reads the skill and follows the instructions.
Skill link
Ready-to-paste message
How it works
It uses a system called 'Model Armor' to analyze text. This system checks if the content follows safety rules before being shown to you or others.
Installation
Easiest way: copy the skill link and paste it into Claude, ChatGPT or Cursor asking it to follow the instructions.
In the terminal (Claude Code, Codex, Cursor), in your project folder:
npx skills add googleworkspace/cli --skill gws-modelarmorOr download just the file:
curl -L https://raw.githubusercontent.com/googleworkspace/cli/main/skills/gws-modelarmor/SKILL.md -o SKILL.mdTechnical content (original SKILL.md)
Original text by the author, straight from GitHub.
modelarmor (v1)
PREREQUISITE: Read
../gws-shared/SKILL.mdfor auth, global flags, and security rules. If missing, rungws generate-skillsto create it.
gws modelarmor <resource> <method> [flags]
Helper Commands
| Command | Description |
|---|---|
+sanitize-prompt |
Sanitize a user prompt through a Model Armor template |
+sanitize-response |
Sanitize a model response through a Model Armor template |
+create-template |
Create a new Model Armor template |
Discovering Commands
Before calling any API method, inspect it:
# Browse resources and methods
gws modelarmor --help
# Inspect a method's required params, types, and defaults
gws schema modelarmor.<resource>.<method>
Use gws schema output to build your --params and --json flags.
FAQ
What is 'Model Armor'?
'Model Armor' is a set of rules and tools that Google uses to ensure artificial intelligence generates safe and responsible content.
Do I need to install anything to use it?
This command is part of a Google toolset. If you already have access to these tools, you can use it. Otherwise, you might need to set them up first.
Can I create my own safety rules?
Yes, 'Model Armor' allows you to create your own 'templates' to define what is safe or not, according to your needs.
Details
Original description by the author
Google Model Armor: Filter user-generated content for safety.
repo: googleworkspace/cli
Popularity
- GitHub stars
- 32k
- Ranking position
- #6278
- License
- Apache-2.0
- Updated
- 10/6/2026
Comments
- …
More skills from this repository
gws-admin-reports
Helps your agent get activity reports from Google Workspace. It collects audit and usage data.
googleworkspace/cli
gws-calendar-agenda
Helps your agent see upcoming events across all calendars. It's useful for organizing your schedule and knowing what's coming up.
googleworkspace/cli
gws-calendar-insert
Lets your agent create new events in Google Calendar. It's useful for scheduling appointments and meetings.
googleworkspace/cli
gws-calendar
Allows your agent to manage calendars and events. It's useful for organizing your life and work.
googleworkspace/cli
Similar skills
bengali-nlp
Has your agent process Bengali text. It helps understand and work with the language, normalizing text and handling its unique characteristics.
affaan-m/ECC
prompt-caching-strategy
Helps your agent optimize how it uses prompts for language models. It analyzes how prompts are made to find ways to cut costs and make responses faster by reusing information when possible.
affaan-m/ECC
prompt-optimizer
Has your agent analyze and improve the prompts you write. It suggests how to add information or change the wording so the AI model understands better and gives the desired answer.
affaan-m/ECC