Create Prompt

Create a new prompt with initial version

Endpoint

method · string · default: "POST"

POST

endpoint · string

/v1/prompt/create

Authentication

This endpoint requires API key authentication. Include your API key in the request headers:

Authorization: Bearer YOUR_API_KEY

Request Body

userDefinedId · string · required

User-defined identifier for the prompt. This is used to reference the prompt in your code

prompt · object · required

The prompt template content. Should follow OpenAI chat completion format with messages array and model

metadata · object · required

Additional metadata for the prompt as key-value pairs

Response

data · object

Created prompt information

id · string

Unique identifier for the prompt

prompt_version_id · string

Unique identifier for the initial prompt version

error · string | null

Error message if the request failed, null otherwise

Example Request

curl -X POST https://api.helicone.ai/v1/prompt/create \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "userDefinedId": "customer-support-v1",
    "prompt": {
      "model": "gpt-4",
      "messages": [
        {
          "role": "system",
          "content": "You are a helpful customer support assistant."
        },
        {
          "role": "user",
          "content": "{{user_question}}"
        }
      ]
    },
    "metadata": {
      "department": "support",
      "version": "1.0"
    }
  }'

Example Response

{
  "data": {
    "id": "prompt_abc123",
    "prompt_version_id": "pv_xyz789"
  },
  "error": null
}

Notes

• The userDefinedId must be unique within your organization
• Use {{variable_name}} syntax for template variables in your prompt
• The initial version is automatically set as version 1.0