sk Skill · googleworkspace
gws-modelarmor-sanitize-prompt
Google Model Armor: Sanitize a user prompt through a Model Armor template.
Open on skills.sh ↗read 2026-09-15
- installs 8w
- 4,246
- 30-day movement
- starts with the next reading
- Related entries
- 2
- Connections
- 0
bashRust
An agent skill for Claude Code that runs a Google Model Armor 'sanitize prompt' operation, filtering user-supplied text through a Model Armor template. It is invoked as modelarmor +sanitize-prompt and bundles a shared gws-modelarmor skill plus a gws-shared skill covering auth and global flags.
You reach for it when you need an AI coding agent to check user prompt text against Google Model Armor safety filters from the command line.
Use it to
- Sanitize a user prompt against a Model Armor template
- Pass prompt text via --text, --, or stdin
- Load auth and global flag rules from gws-shared before running
- Pair with +sanitize-response for outbound safety checks
For Claude Code users working with Google Workspace Model Armor
- Host repository
- googleworkspace/cli
- Version
- 0.22.5
- Installs, lifetime
- 27k
- Installs, 8 weeks
- 4,246
- Host stars
- 31k
- Host language
- Rust
topicsmodel-armorgoogle-workspaceprompt-sanitizationskillsecuritybash