A digital scene in classic 8-bit style shows stacked rectangles with tabs, symbolizing documents, linked by bold arrows to a computer screen with bracket and brace icons representing code. Small circles suggest variables, a simple clock icon refers to time, and minimal template shapes are present. The image uses five corporate colors on a flat, single-color background, is highly simplified and geometric, and measures 128 by 128 pixels. There are no people, no words, and no realistic effects.

Import Template

Various LLM tools allow storing prompts in text or markdown files. You can use importTemplate to import these files into a prompt.

Explain your answer step by step.

importTemplate("cot.md")

Variable interpolation

importTemplate supports mustache (default), Jinja variable interpolation and the Prompty file format. You can use variables in the imported template and pass them as arguments to the importTemplate function.

The current time is {{time}}.

importTemplate("time.md", { time: "12:00" })

Mustache supports arguments as functions. This allows you to pass dynamic values to the template.

importTemplate("time.md", { time: () => Date.now() })

More way to specify files

You can use the results of workspace.readText.

const file = await workspace.readText("time.md")
importTemplate(time, { time: "12:00" })

You can specify an array of files or glob patterns.

importTemplate("*.prompt")

Prompty

Prompty provides a simple markdown-based format for prompts. It adds the concept of role sections to the markdown format.

---
name: Basic Prompt
description: A basic prompt that uses the chat API to answer questions
---

inputs:
question:
type: string
sample:
"question": "Who is the most famous person in the world?"

---

system:
You are an AI assistant who helps people find information.
As the assistant, you answer questions briefly, succinctly.

user:
{{question}}

importTemplate("basic.prompty", { question: "what is the capital of France?" })