Create a Guardrail - Orq.ai Documentation

POST

evaluators

Create an Evaluator

curl --request POST \
  --url https://api.orq.ai/v2/evaluators \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "type": "llm_eval",
  "prompt": "<string>",
  "path": "Default",
  "key": "<string>",
  "mode": "single",
  "model": "<string>",
  "guardrail_config": {
    "type": "boolean",
    "value": true,
    "enabled": true,
    "alert_on_failure": false
  },
  "repetitions": 2,
  "categories": [
    "<string>"
  ],
  "categorical_labels": [
    {
      "value": "<string>",
      "description": "<string>"
    }
  ],
  "dataset_id": "<string>",
  "description": ""
}
'

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    type: 'llm_eval',
    prompt: '<string>',
    path: 'Default',
    key: '<string>',
    mode: 'single',
    model: '<string>',
    guardrail_config: {type: 'boolean', value: true, enabled: true, alert_on_failure: false},
    repetitions: 2,
    categories: ['<string>'],
    categorical_labels: [{value: '<string>', description: '<string>'}],
    dataset_id: '<string>',
    description: ''
  })
};

fetch('https://api.orq.ai/v2/evaluators', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

import requests

url = "https://api.orq.ai/v2/evaluators"

payload = {
    "type": "llm_eval",
    "prompt": "<string>",
    "path": "Default",
    "key": "<string>",
    "mode": "single",
    "model": "<string>",
    "guardrail_config": {
        "type": "boolean",
        "value": True,
        "enabled": True,
        "alert_on_failure": False
    },
    "repetitions": 2,
    "categories": ["<string>"],
    "categorical_labels": [
        {
            "value": "<string>",
            "description": "<string>"
        }
    ],
    "dataset_id": "<string>",
    "description": ""
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "_id": "<string>",
  "description": "<string>",
  "type": "llm_eval",
  "prompt": "<string>",
  "key": "<string>",
  "mode": "single",
  "model": "<string>",
  "created": "2026-06-10T10:08:15.342Z",
  "updated": "2026-06-10T10:08:15.342Z",
  "updated_by_id": "<string>",
  "guardrail_config": {
    "type": "boolean",
    "value": true,
    "enabled": true,
    "alert_on_failure": false
  },
  "repetitions": 2,
  "categories": [
    "<string>"
  ],
  "categorical_labels": [
    {
      "value": "<string>",
      "description": "<string>"
    }
  ],
  "dataset_id": "<string>"
}

{
  "message": "<string>"
}

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

LLM
LLM
Python

type

enum<string>

required

Available options:

llm_eval

prompt

string

required

path

string

required

Entity storage path.

With workspace-level API keys, use the format project/folder/subfolder/.... The first element identifies the project, followed by nested folders (auto-created as needed). Example: Default/agents.

With project-level API keys, the project is predetermined by the API key, so the path is relative to that project. Example: agents. For backward compatibility, a leading project name is ignored when it matches the scoped project.

Example:

"Default"

key

string

required

Minimum string length: 1

Pattern: ^[a-zA-Z0-9]([a-zA-Z0-9_-]*[a-zA-Z0-9])?$

mode

enum<string>

required

Available options:

single

model

string

required

guardrail_config

Boolean · object

Boolean
Categorical
Number

Show child attributes

output_type

enum<string>

The type of output expected from the evaluator

Available options:

boolean,

categorical,

number,

string

repetitions

integer | null

Required range: 1 <= x <= 3

Response

Successfully created an evaluator

_id

string

required

description

string

required

type

enum<string>

required

Available options:

llm_eval

prompt

string

required

key

string

required

Minimum string length: 1

Pattern: ^[a-zA-Z0-9]([a-zA-Z0-9_-]*[a-zA-Z0-9])?$

mode

enum<string>

required

Available options:

single

model

string

required

created

string

default:2026-06-10T10:08:15.342Z

updated

string

default:2026-06-10T10:08:15.342Z

updated_by_id

string | null

guardrail_config

Boolean · object

Boolean
Categorical
Number

Show child attributes

repetitions

integer | null

Required range: 1 <= x <= 3

categories

string[] | null

categorical_labels

object[] | null

Show child attributes

dataset_id

string

Get all Guardrails

Delete a Guardrail

⌘I

Create an Evaluator

curl --request POST \
  --url https://api.orq.ai/v2/evaluators \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "type": "llm_eval",
  "prompt": "<string>",
  "path": "Default",
  "key": "<string>",
  "mode": "single",
  "model": "<string>",
  "guardrail_config": {
    "type": "boolean",
    "value": true,
    "enabled": true,
    "alert_on_failure": false
  },
  "repetitions": 2,
  "categories": [
    "<string>"
  ],
  "categorical_labels": [
    {
      "value": "<string>",
      "description": "<string>"
    }
  ],
  "dataset_id": "<string>",
  "description": ""
}
'

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    type: 'llm_eval',
    prompt: '<string>',
    path: 'Default',
    key: '<string>',
    mode: 'single',
    model: '<string>',
    guardrail_config: {type: 'boolean', value: true, enabled: true, alert_on_failure: false},
    repetitions: 2,
    categories: ['<string>'],
    categorical_labels: [{value: '<string>', description: '<string>'}],
    dataset_id: '<string>',
    description: ''
  })
};

fetch('https://api.orq.ai/v2/evaluators', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

import requests

url = "https://api.orq.ai/v2/evaluators"

payload = {
    "type": "llm_eval",
    "prompt": "<string>",
    "path": "Default",
    "key": "<string>",
    "mode": "single",
    "model": "<string>",
    "guardrail_config": {
        "type": "boolean",
        "value": True,
        "enabled": True,
        "alert_on_failure": False
    },
    "repetitions": 2,
    "categories": ["<string>"],
    "categorical_labels": [
        {
            "value": "<string>",
            "description": "<string>"
        }
    ],
    "dataset_id": "<string>",
    "description": ""
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "_id": "<string>",
  "description": "<string>",
  "type": "llm_eval",
  "prompt": "<string>",
  "key": "<string>",
  "mode": "single",
  "model": "<string>",
  "created": "2026-06-10T10:08:15.342Z",
  "updated": "2026-06-10T10:08:15.342Z",
  "updated_by_id": "<string>",
  "guardrail_config": {
    "type": "boolean",
    "value": true,
    "enabled": true,
    "alert_on_failure": false
  },
  "repetitions": 2,
  "categories": [
    "<string>"
  ],
  "categorical_labels": [
    {
      "value": "<string>",
      "description": "<string>"
    }
  ],
  "dataset_id": "<string>"
}

{
  "message": "<string>"
}