Text Prompt

The default prompt is a text prompt, and is meant for models such as xx and xx.

You create a prompt using createPrompt().

// create a simple text prompt
const prompt = createPrompt("text")

{
  "type": "text",
  "messages": [],
  "partials": [],
  "helpers": []
}

Hide Log

When creating a prompt, you can optionally set an initial message.

const initial = "You are a customer service agent for SomeCompany.";
const prompt = createPrompt("text", initial)

{
  "type": "text",
  "messages": [
    {
      "role": "system",
      "content": "You are a customer service agent for SomeCompany."
    }
  ],
  "partials": [],
  "helpers": []
}

Hide Log

You can also add additional content to the prompt by calling addToPrompt() on the prompt.

// create a simple text prompt with an initial system message
const initial = "You are a customer service agent for SomeCompany.";
const prompt = createPrompt("text", initial)

// add an additional system message to the prompt
prompt.addToPrompt("You should never admit that you are AI.");

{
  "type": "text",
  "messages": [
    {
      "role": "system",
      "content": "You are a customer service agent for SomeCompany."
    },
    {
      "role": "system",
      "content": "You should never admit that you are AI."
    }
  ],
  "partials": [],
  "helpers": []
}

Hide Log

To use the prompt as input to an LLM, you can call the format() method on the prompt. The format method accepts an object, which is used to supply the prompt template with replacement values.

// create a simple text prompt with an initial system message
const initial = "You are a customer service agent for SomeCompany.";
const prompt = createPrompt("text", initial)

// add an additional system message to the prompt
prompt.addToPrompt("You should never admit that you are AI.");

// format the prompt with values
const formatted = prompt.format({});

You are a customer service agent for SomeCompany.

You should never admit that you are AI.

Hide Log

Prompt methods are chainable

const exampleCreateTextPromptChainable = () => {
const initial = "You are a customer service agent for SomeCompany.";
const prompt = createPrompt("text", initial)
    
// you can also chain all prompt methods (except format)
prompt
.addToPrompt("You should never admit that you are AI.")
.addToPrompt("Begin!")

{
  "type": "text",
  "messages": [
    {
      "role": "system",
      "content": "You are a customer service agent for SomeCompany."
    },
    {
      "role": "system",
      "content": "You should never admit that you are AI."
    },
    {
      "role": "system",
      "content": "Begin!"
    }
  ],
  "partials": [],
  "helpers": []
}

Hide Log

By default, formatted text prompt messages are separated using 2 line breaks (\n\n). You can override this by defining a custom separator.

const initial = "You are a customer service agent for SomeCompany.";
const prompt = createPrompt("text", initial)
.addToPrompt("You should never admit that you are AI.")
.addToPrompt("Begin!")
    
// you can also define a custom separator between the messages
const withCustomSeparator = prompt.format({}, "\n---\n")     

You are a customer service agent for SomeCompany.
---
You should never admit that you are AI.
---
Begin!

Hide Log

See prompt templates for more advanced prompt usage.

Text Prompt Methods

.addToPrompt() Adds content to the prompt. @param content {string} The content to be added to the prompt.

.registerPartial()partials {<{template: string; name: string;}>} Additional partials that can be made available to the template parser.

.registerHelpers()helpers {<{handler: function; name: string;}>} Additional helper functions that can be made available to the template parser.

.format()format Processes the prompt template and returns prompt ready for LLM.