gws-modelarmor

A command set for Google Model Armor, a Google Cloud service that checks text for safety risks. It can create filtering templates and sanitize user prompts or model responses with them.

In plain words
What is it for?
Use it to inspect Model Armor commands, create templates, and filter incoming prompts or outgoing model responses through a chosen template.
Why use it?
It helps screen text before it reaches an AI model or a user. This reduces the need to build these safety checks into each application yourself.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/googleworkspace/cli/gws-modelarmor
Any agent
npx skills add googleworkspace/cli --skill gws-modelarmor
Clone the repo
git clone --depth 1 https://github.com/googleworkspace/cli

Made for: Claude Code, Codex.

Per session 16 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 321 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00016 $0.00321
Opus 5 $0.00008 $0.00161
Sonnet 5 $0.00003 $0.00064
Haiku 4.5 $0.00002 $0.00032

Measured 2d ago against content hash 072890561fcd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

gws-modelarmor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

3 near-identical copies found in the catalogue:

skills/gws-modelarmor/SKILL.md · 44 lines

What it actually says

modelarmor (v1)

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

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 44 lines · 16 tokens per session scan A 072890561fcd

Subscribe to this mod's changes

gws-modelarmor is a skill published in the GitHub repository googleworkspace/cli (30,691 stars, last pushed 7d ago), licensed Apache-2.0. It adds 16 tokens to every session and 321 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.