gws-modelarmor-sanitize-prompt

1# modelarmor +sanitize-prompt
2 
3> **PREREQUISITE:** Read `../gws-shared/SKILL.md` for auth, global flags, and security rules. If missing, run `gws generate-skills` to create it.
4 
5Sanitize a user prompt through a Model Armor template
6 
7## Usage
8 
9```bash
10gws modelarmor +sanitize-prompt --template <NAME>
11```
12 
13## Flags
14 
15| Flag | Required | Default | Description |
16|------|----------|---------|-------------|
17| `--template` | ✓ | — | Full template resource name (projects/PROJECT/locations/LOCATION/templates/TEMPLATE) |
18| `--text` | — | — | Text content to sanitize |
19| `--json` | — | — | Full JSON request body (overrides --text) |
20 
21## Examples
22 
23```bash
24gws modelarmor +sanitize-prompt --template projects/P/locations/L/templates/T --text 'user input'
25echo 'prompt' | gws modelarmor +sanitize-prompt --template ...
26```
27 
28## Tips
29 
30- If neither --text nor --json is given, reads from stdin.
31- For outbound safety, use +sanitize-response instead.
32 
33## See Also
34 
35- [gws-shared](../gws-shared/SKILL.md) — Global flags and auth
36- [gws-modelarmor](../gws-modelarmor/SKILL.md) — All filter user-generated content for safety commands