Create condition

POST /automation/conditions/

Create Condition

cURL

curl --request POST \
  --url https://api.keywordsai.co/automation/conditions/ \
  --header 'Authorization: Bearer <token>'

{
  "id": "cond-12345",
  "name": "Successful Requests",
  "condition_slug": "success_logs",
  "condition_type": "single_log",
  "description": "Matches logs with HTTP status code 200",
  "condition_policy": {
    "rules": [
      {
        "field": "status_code",
        "operator": "equals",
        "value": 200
      }
    ],
    "connector": "AND"
  },
  "unique_organization_id": "org-abc123",
  "created_at": "2025-01-15T10:00:00Z",
  "updated_at": "2025-01-15T10:00:00Z"
}

Creates a new automation condition that defines which logs should trigger evaluations.

Authentication

All endpoints require API key authentication:

Authorization: Bearer YOUR_API_KEY

Note: Use your API Key (not JWT token) for all requests. You can find your API keys in the Keywords AI platform under Settings > API Keys.

Required Fields

Field	Type	Description
name	string	Display name for the condition
condition_slug	string	Unique identifier for the condition
condition_type	string	Must be "single_log" for online eval
condition_policy	object	JSON structure defining evaluation rules

Optional Fields

Field	Type	Description
description	string	Description of the condition

Condition Policy Structure

The condition_policy object contains:

• rules (array): Array of rule objects, each with:
  • field (string): Field to evaluate (e.g., "status_code", "latency", "model")
  • operator (string): Comparison operator (see available operators below)
  • value (any): Value to compare against
• connector (string): How to combine rules ("AND" or "OR")

Available Operators

• equals, not_equals
• greater_than, less_than, greater_than_or_equal, less_than_or_equal
• contains, not_contains
• in, not_in

Available Fields

• status_code: HTTP status code
• latency: Request latency in seconds
• total_tokens: Total token count
• input_tokens: Input token count
• output_tokens: Output token count
• cost: Request cost
• model: Model name
• Any custom metadata field

Examples

Single Rule Condition

Python

import requests

url = "https://api.keywordsai.co/automation/conditions/"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}

data = {
    "name": "Successful Requests",
    "condition_slug": "success_logs",
    "condition_type": "single_log",
    "description": "Matches logs with HTTP status code 200",
    "condition_policy": {
        "rules": [
            {
                "field": "status_code",
                "operator": "equals",
                "value": 200
            }
        ],
        "connector": "AND"
    }
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

Multiple Rules Condition

Python

data = {
    "name": "High Latency Errors",
    "condition_slug": "high_latency_errors",
    "condition_type": "single_log",
    "condition_policy": {
        "rules": [
            {
                "field": "status_code",
                "operator": "greater_than_or_equal",
                "value": 500
            },
            {
                "field": "latency",
                "operator": "greater_than",
                "value": 2.0
            }
        ],
        "connector": "AND"
    }
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

Condition with Custom Metadata Field

Python

data = {
    "name": "VIP Customer Requests",
    "condition_slug": "vip_customer_logs",
    "condition_type": "single_log",
    "condition_policy": {
        "rules": [
            {
                "field": "metadata.customer_tier",
                "operator": "equals",
                "value": "VIP"
            },
            {
                "field": "status_code",
                "operator": "equals",
                "value": 200
            }
        ],
        "connector": "AND"
    }
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

Response

Status: 201 Created

{
  "id": "cond-12345",
  "name": "Successful Requests",
  "condition_slug": "success_logs",
  "condition_type": "single_log",
  "description": "Matches logs with HTTP status code 200",
  "condition_policy": {
    "rules": [
      {
        "field": "status_code",
        "operator": "equals",
        "value": 200
      }
    ],
    "connector": "AND"
  },
  "unique_organization_id": "org-abc123",
  "created_at": "2025-01-15T10:00:00Z",
  "updated_at": "2025-01-15T10:00:00Z"
}

Response Fields

Field	Type	Description
id	string	Unique condition identifier (save this for creating automations)
name	string	Display name of the condition
condition_slug	string	URL-friendly identifier
condition_type	string	Type of condition ("single_log")
description	string	Description of the condition
condition_policy	object	The condition policy with rules and connector
unique_organization_id	string	Organization identifier
created_at	string	ISO timestamp of creation
updated_at	string	ISO timestamp of last update

Error Responses

400 Bad Request

{
  "condition_policy": [
    "Invalid operator 'invalid_op' for field 'status_code'"
  ]
}

401 Unauthorized

{
  "detail": "Authentication credentials were not provided."
}

422 Unprocessable Entity

{
  "condition_type": [
    "Condition type must be 'single_log' for online evaluation automations"
  ]
}