> ## Documentation Index
> Fetch the complete documentation index at: https://apie.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Enable guardrail templates

> Turn on starter guardrail packs from the Apie dashboard via the CLI.

Apie ships starter guardrail templates — pre-built policies for common production risks like secret access, production deploys, and destructive shell commands. Enable them with a single CLI command before switching to Enforce mode.

When you finish this page, guardrail templates will be active for your agent in the dashboard.

## Enable a template

<CodeGroup>
  ```bash TypeScript theme={null}
  npx apie guardrails enable prod-secrets
  npx apie guardrails enable prod-deploys
  ```

  ```bash Python theme={null}
  apie guardrails enable prod-secrets
  apie guardrails enable prod-deploys
  ```
</CodeGroup>

You can also enable templates programmatically:

<CodeGroup>
  ```ts TypeScript theme={null}
  await apie.enableGuardrailTemplate("prod-secrets");
  ```

  ```python Python theme={null}
  apie.enable_guardrail_template("prod-secrets")
  ```
</CodeGroup>

## Starter packs

These templates match common risky action patterns. Exact keys depend on your Apie workspace — check the dashboard for available templates.

| Template key    | Typically guards                            |
| --------------- | ------------------------------------------- |
| `prod-secrets`  | Secret and vault reads in production        |
| `prod-deploys`  | Deployment and release executions           |
| `prod-database` | Database writes and schema changes          |
| `prod-shell`    | Destructive shell commands (`rm -rf`, etc.) |
| `prod-merge`    | Production branch merges                    |

## Test templates in monitor mode

Before enforcing, exercise templates in monitor mode to see which decisions they produce:

```bash theme={null}
npx apie send-test-event --mode pipeline
```

Or run the [Guardrail packs recipe](/recipes/guardrail-packs) which exercises deploy, vault, database, merge, and shell scenarios.

### What you'll see

`agent.guardrail.evaluated` events showing which templates matched and what decision they returned (`allow`, `warn`, `block`, `require_approval`).

## Workflow

1. Enable templates via CLI
2. Run in [monitor mode](/guardrails/monitor-mode) and review evaluations
3. [Declare capabilities](/boundaries/declare-capabilities) so templates have accurate context
4. Switch to [Enforce mode](/guardrails/enforce-guardrails)

## Next steps

<CardGroup cols={2}>
  <Card title="Enforce guardrails" icon="shield" href="/guardrails/enforce-guardrails">
    Block actions that templates flag.
  </Card>

  <Card title="Guardrail packs recipe" icon="flask" href="/recipes/guardrail-packs">
    Walk through all five smoke scenarios.
  </Card>
</CardGroup>