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

# Guardrail Reference

> Reference for guardrail entities, statuses, and key operations

## Core Entities

| Entity     | Purpose                          | Key Fields                          |
| ---------- | -------------------------------- | ----------------------------------- |
| Provider   | Source of probe catalog          | name, type, availability            |
| Probe      | Detection capability unit        | id, tags, modality, guard types     |
| Rule       | Fine-grained detector in a probe | id, status, threshold override      |
| Profile    | Reusable guardrail policy        | name, probes, guard types, severity |
| Deployment | Runtime activation of profile    | project, endpoint, status           |

## Status Reference

| Status                  | Meaning                                  |
| ----------------------- | ---------------------------------------- |
| `active`                | Currently enabled and protecting traffic |
| `disabled`              | Configured but not enforcing             |
| `pending` / `deploying` | Deployment workflow in progress          |
| `deleted`               | Removed from active usage                |
| `unhealthy` / `failure` | Requires investigation and remediation   |

## Guardrail Workflow Reference

```mermaid theme={null}
flowchart LR
    A[Select Provider] --> B[Select Probes]
    B --> C[Configure Profile]
    C --> D[Choose Deployment Type]
    D --> E[Select Project/Endpoint]
    E --> F[Deploy]
```

## Common Operations

### Probe Discovery

* List probes by provider and tag.
* Inspect probe details and rule coverage.

### Profile Management

* Create profile with selected probes.
* Configure guard types and severity.
* Update profile over time with controlled rollout.

### Deployment Management

* Deploy profile to project/endpoint.
* Monitor status.
* Reconfigure or remove deployment as needed.

## Configuration Guidance

| Configuration    | Recommendation                          |
| ---------------- | --------------------------------------- |
| Initial severity | Start medium, tune with real traffic    |
| Probe count      | Start minimal, then expand deliberately |
| Rollout strategy | Pilot first, then scale by project      |
| Documentation    | Record owner, intent, and exceptions    |