Skip to main content
GET
/
prompt-templates
/
{identifier}
Get Prompt Template Raw Data
curl --request GET \
  --url https://api.promptlayer.com/prompt-templates/{identifier} \
  --header 'X-API-KEY: <x-api-key>'
{
  "success": true,
  "id": 123,
  "prompt_name": "<string>",
  "version": 123,
  "workspace_id": 123,
  "prompt_template": {
    "content": [
      {
        "text": "<string>",
        "type": "text"
      }
    ],
    "input_variables": [],
    "template_format": "f-string",
    "type": "completion"
  },
  "snippets": [
    {
      "prompt_name": "<string>",
      "version": 123,
      "label": "<string>"
    }
  ],
  "metadata": {
    "model": {
      "provider": "<string>",
      "name": "<string>",
      "parameters": {}
    },
    "customField": "<string>"
  },
  "commit_message": "<string>",
  "tags": [
    "<string>"
  ],
  "created_at": "2023-11-07T05:31:56Z",
  "llm_kwargs": {}
}
Retrieve raw prompt template data without applying input variables. This endpoint is designed for:
  • GitHub sync: Use resolve_snippets=false to get the raw template with @@@snippet@@@ references intact
  • Local caching: Fetch resolved templates with optional llm_kwargs for offline use
  • Template inspection: View template structure and metadata without executing
Unlike the POST endpoint, this GET endpoint does not accept input_variables or provider in a request body. Instead, it returns the raw template data with placeholders preserved.

​
Query Parameters

ParameterTypeDefaultDescription
versioninteger-Specific version number. Mutually exclusive with label.
labelstring-Release label name (e.g. prod). Mutually exclusive with version.
resolve_snippetsbooleantrueWhen true, snippets are expanded. When false, raw @@@snippet@@@ references are preserved.
include_llm_kwargsbooleanfalseWhen true, includes provider-specific LLM API format in the response.

​
Caching

Responses are cached by default. To bypass the cache, send the Cache-Control: no-cache header.
Provider-Specific Schema NoticeThe llm_kwargs field (when requested via include_llm_kwargs=true) is provider-specific and its structure may change without notice as LLM providers update their APIs.For stable, provider-agnostic prompt data, use prompt_template instead of llm_kwargs.

​
Examples

# Default: resolved snippets, latest version
curl -H "X-API-KEY: your_api_key" \
  https://api.promptlayer.com/prompt-templates/my-prompt

# Raw template with snippet references (for GitHub sync)
curl -H "X-API-KEY: your_api_key" \
  "https://api.promptlayer.com/prompt-templates/my-prompt?resolve_snippets=false"

# With llm_kwargs (for local caching)
curl -H "X-API-KEY: your_api_key" \
  "https://api.promptlayer.com/prompt-templates/my-prompt?include_llm_kwargs=true"

# Specific version
curl -H "X-API-KEY: your_api_key" \
  "https://api.promptlayer.com/prompt-templates/my-prompt?version=2"

# By release label
curl -H "X-API-KEY: your_api_key" \
  "https://api.promptlayer.com/prompt-templates/my-prompt?label=prod"

​
Authentication

This endpoint requires API key authentication via the X-API-KEY header.

Headers

​
X-API-KEY
string
required

Path Parameters

​
identifier
string
required

The identifier can be either the prompt name or the prompt id.

Query Parameters

​
version
integer

Specific version number to retrieve. Mutually exclusive with label.

​
label
string

Release label name to retrieve (e.g. 'prod', 'staging'). Mutually exclusive with version.

​
resolve_snippets
boolean
default:true

When true (default), snippets are expanded in the returned prompt_template. When false, raw @@@snippet@@@ references are preserved.

​
include_llm_kwargs
boolean
default:false

When true, includes provider-specific llm_kwargs in the response. Requires model metadata to be set on the template.

Response

Successful Response

​
success
boolean
required
​
id
integer
required

The prompt template ID.

​
prompt_name
string
required

The name of the prompt template.

​
version
integer
required

The version number of the prompt template.

​
workspace_id
integer
required

The workspace this prompt template belongs to.

​
prompt_template
Completion Template · object
required

The prompt template content. When resolve_snippets is true (default), snippets are expanded. When false, raw @@@snippet@@@ references are preserved.

​
snippets
SnippetReference · object[]
required

List of snippet references used in this template.

​
metadata
Metadata · object

Model configuration including provider, model name, and parameters.

​
commit_message
string | null

The commit message for this version.

​
tags
string[]

Tags associated with the prompt template.

​
created_at
string<date-time> | null

Timestamp when this version was created.

​
llm_kwargs
LLM Kwargs · object

Provider-specific LLM arguments. Only present when include_llm_kwargs=true. Structure is provider-specific and may change without notice.