Gws Modelarmor Sanitize Prompt
- 23.6k installs
- 30k repo stars
- Updated July 22, 2026
- googleworkspace/cli
gws-modelarmor-sanitize-prompt is a CLI skill that filters user prompts through Model Armor templates.
About
CLI command to filter user prompts through Google Model Armor safety templates before processing by LLM. Developers use this for inbound content safety - preventing prompt injection and malicious user input. Complements sanitize-response for comprehensive LLM safety.
- Sanitizes user prompts through Model Armor templates
- Inbound safety filtering for user-generated content
- Supports stdin, text flag, and JSON request body customization
Gws Modelarmor Sanitize Prompt by the numbers
- 23,643 all-time installs (skills.sh)
- +799 installs in the week ending Jul 28, 2026 (Skillselion tracking)
- Ranked #22 of 2,209 Security skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Jul 28, 2026 (Skillselion catalog sync)
gws-modelarmor-sanitize-prompt capabilities & compatibility
- Works with
- gcp
- Use cases
- security audit
What gws-modelarmor-sanitize-prompt says it does
Sanitize a user prompt through a Model Armor template
For outbound safety, use +sanitize-response instead
If neither --text nor --json is given, reads from stdin
npx skills add https://github.com/googleworkspace/cli --skill gws-modelarmor-sanitize-promptAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 23.6k |
|---|---|
| repo stars | ★ 30k |
| Security audit | 3 / 3 scanners passed |
| Last updated | July 22, 2026 |
| Repository | googleworkspace/cli ↗ |
How do you sanitize prompts before LLM inference?
Filter user prompts through Model Armor to prevent injection attacks before processing.
Who is it for?
Preventing prompt injection and filtering user input before LLM processing
Skip if: Output filtering after generation or teams still needing to create a Model Armor template first.
When should I use this skill?
Need to validate/filter user-generated prompts against jailbreak attempts
What you get
A Model Armor-sanitized prompt cleared or flagged by the configured safety template.
- Filtered/sanitized prompt text
- Safety validation result
By the numbers
- Metadata version 0.22.5 in googleworkspace/cli
- Documents gws modelarmor +sanitize-prompt with required --template flag
Files
modelarmor +sanitize-prompt
PREREQUISITE: Read../gws-shared/SKILL.mdfor auth, global flags, and security rules. If missing, rungws generate-skillsto 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 ...Tips
- If neither --text nor --json is given, reads from stdin.
- For outbound safety, use +sanitize-response instead.
See Also
- gws-shared — Global flags and auth
- gws-modelarmor — All filter user-generated content for safety commands
Related skills
Entra App RegistrationCorrectly register an application in Microsoft Entra ID, configure OAuth 2.0 flows, request the right API permissions, and generate working MSAL authentication snippets476k1.3kAzure ComplianceRun automated Azure compliance scans, security posture checks, and Key Vault expiration audits before deploying.475k1.3k
Openclaw Secure Linux CloudDeploy and harden an OpenClaw agent instance on a Linux cloud server following battle-tested security defaults.270k72
Better Auth Best PracticesCorrectly configure Better Auth for secure authentication with database adapters, sessions, OAuth, email/password, and plugins in TypeScript projects.78.9k204
Firebase Security Rules AuditorAutomatically audit Firebase Firestore security rules for bypass vulnerabilities and logic gaps before deploying.77.7k388
Audit WebsiteRun comprehensive audits that surface SEO, performance, security, technical, and content issues with LLM-optimized reports and health scores.64.9k85How it compares
Use gws-modelarmor-sanitize-prompt for inbound prompt checks; use gws-modelarmor sanitize-response helpers for post-generation output filtering.
FAQ
What happens to unsafe prompts?
Model Armor returns a filtered version or rejection - implementation depends on template configuration.
Can I use stdin?
Yes - if neither --text nor --json is provided, the command reads from stdin (e.g., `echo prompt | gws modelarmor +sanitize-prompt ...`)