Discover how to get your prompt. You can do it with a single line of code or specify a version or tag.

​
Single line of code

You can get your prompt with a single line of code.

const { value } await basalt.prompt.get('my-prompt-slug')

​
Get a specific Tag

You can retrieve a specific tag of a prompt by providing the tag parameter.

// Using object syntax
const { value } = await basalt.prompt.get({
  slug: 'welcome-message',
  tag: 'staging'
})

// Using parameter syntax
const { value } = await basalt.prompt.get('welcome-message', {
  tag: 'staging'
})

If both version and tag are provided, version takes precedence. If neither is provided, the system attempts to find the production tagged version

​
Get a specific Version

You can retrieve a specific version of a prompt by providing the version parameter.

// Using object syntax
const { value } = await basalt.prompt.get({
  slug: 'welcome-message',
  version: '1.0.0'
})

// Using parameter syntax
const { value } = await basalt.prompt.get('welcome-message', {
  version: '1.0.0'
})

​
Replace Variables

If your prompt contains variables (e.g., {{name}}), you can replace them with actual values when retrieving the prompt:

const { value } = await basalt.prompt.get({
  slug: 'welcome-message',
  variables: {
    name: 'John Doe',
    company: 'Acme Inc'
  }
})

// The variables can also be provided with the parameter syntax
const { value } = await basalt.prompt.get('welcome-message', {
  variables: {
    name: 'John Doe'
  }
})

​
Returned data

When you retrieve a prompt, the SDK returns the prompt (user and system) along with its model configuration.

interface PromptResult {
	text: string; // User Prompt
	systemText: string | undefined; // System Prompt
	model: {
		provider: 'anthropic' | 'open-ai' | 'mistral' | 'gemini';
		model: string;
		version: string | 'latest';
		parameters: {
			temperature: number;
			topP: number;
			frequencyPenalty?: number;
			presencePenalty?: number;
			topK?: number;
			maxLength: number;
			responseFormat: ResponseFormat;
			jsonObject?: Record<string, any>;
		};
	};
}

​
Example response

Response example
{
  "text": "Can you help me with...",
  "systemText": "You are an AI assistant that helps users with their tasks.",
  "model": {
    "provider": "open-ai",
    "model": "4.1-nano",
    "version": "latest",
    "parameters": {
      "temperature": 0.7,
      "topP": 0.95,
      "frequencyPenalty": 1,
      "presencePenalty": 1,
      "topK": 40,
      "maxLength": 1000,
      "responseFormat": "text",
      "jsonObject": null
    }
  }
}

​
Error handling

Discover how to properly handle errors comming from the SDK and/or the API by checking the Error Handling documentation

​
Caching