List adoptable guardrail presets.

GET

/v1/capabilities/guardrails/examples

const url = 'https://app.everruns.com/api/v1/capabilities/guardrails/examples';
const options = {method: 'GET'};

try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}

curl --request GET \
  --url https://app.everruns.com/api/v1/capabilities/guardrails/examples

Responses

200

Adoptable guardrail presets with trust metadata

application/json

Response for the list_guardrail_examples operation.

object

examples

required

Adoptable guardrail presets, in display order.

Array<object>

A read-only, adoptable guardrails preset from the gallery. Adopt by dropping config into an agent’s guardrails capability config.

object

check_types

required

Distinct rule types used across the preset’s checks (regex, blocklist, tool_pattern) — the check-type composition.

Array<string>

config

required

The adoptable GuardrailsConfig. Same shape persisted in AgentCapabilityConfig.config for the guardrails capability.

object

data_egress

required

Where the preset sends data when it runs. none for deterministic presets (everything runs in-process).

string

description

required

What the preset protects against and how to tune it.

string

display_name

required

Human-facing label.

string

name

required

Stable slug used to reference the preset (e.g. secret-detection).

string

stages

required

Distinct stages the preset’s checks run in (output, tool_use, tool_output).

Array<string>

tags

required

Tags for grouping/filtering in a picker.

Array<string>

Example

{
  "examples": [
    {
      "data_egress": "none",
      "display_name": "Secret & Credential Detection",
      "name": "secret-detection"
    }
  ]
}