Skip to main content
POST
https://api.keywordsai.co
/
automation
/
conditions
Create Condition
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

FieldTypeDescription
namestringDisplay name for the condition
condition_slugstringUnique identifier for the condition
condition_typestringMust be "single_log" for online eval
condition_policyobjectJSON structure defining evaluation rules

​
Optional Fields

FieldTypeDescription
descriptionstringDescription 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

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

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

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

FieldTypeDescription
idstringUnique condition identifier (save this for creating automations)
namestringDisplay name of the condition
condition_slugstringURL-friendly identifier
condition_typestringType of condition ("single_log")
descriptionstringDescription of the condition
condition_policyobjectThe condition policy with rules and connector
unique_organization_idstringOrganization identifier
created_atstringISO timestamp of creation
updated_atstringISO 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"
  ]
}