Prompt templating is a powerful feature that allows you to create dynamic prompts with customizable inputs. This flexibility enables you to design interactive and personalized AI experiences for your users. In this guide, we’ll explore how to use prompt templates effectively in your Aikeedo projects.
Understanding Prompt Templates
Prompt templates are pre-designed structures that include placeholders for user inputs. These templates allow you to create consistent and customizable prompts for various AI-powered tasks, such as content generation, language translation, or data analysis.
Key Components
Tags are the building blocks of prompt templates. They represent input elements that users can customize and are enclosed in curly braces.
Attributes
Attributes modify the behavior of tags, allowing you to fine-tune how inputs are presented and processed. They are appended to tags using the pipe character (|).
{ input_name | attribute1 | attribute2 }
Creating Your First Prompt Template
Let’s create a simple prompt template for an “Article Generator”:
Write an article about { topic | placeholder: Your chosen subject }.
Use the following draft content:
---
{ content | multiline | placeholder: Draft content }
---
Additional instructions:
- Include relevant statistics and consider diverse perspectives.
- Write it in { language } using a { tone } voice tone.
Available Attributes
Enhance your prompt templates with these powerful attributes:
Sets a custom label for the inputExample:
{ input_name | label:Custom label }
Creates a multiline text inputExample:{ input_name | multiline }
Defines the input type (currently supports enum)Example:{ input_name | type:enum }
Specifies options for enum type inputsExample:{ input_name | type:enum | options:Option 1, Option 2, Option 3 }
Makes an input optionalExample:{ input_name | optional }
Sets a custom placeholder textExample:{ input_name | placeholder:Custom placeholder }
Provides additional information about the inputExample:{ input_name | info:Custom info text }
language: A dropdown list of supported languages.tone: A dropdown list of supported voice tones.
Use these built-in inputs in your templates like this:
Advanced Template Example
Let’s create a more sophisticated prompt template for a “Comprehensive Article Generator”:
Create a detailed article on { topic | placeholder: Your chosen subject | info: Specify the main focus of your article }.
Target audience: { audience | type:enum | options:General, Beginners, Experts | info: Select your intended readership }
Article structure:
1. Introduction
2. { section_1 | label:Main Section 1 | placeholder:E.g., Historical background }
3. { section_2 | label:Main Section 2 | placeholder:E.g., Current trends }
4. { section_3 | label:Main Section 3 | placeholder:E.g., Future implications }
5. Conclusion
Key points to cover:
{ key_points | multiline | placeholder: List important points to address }
Writing style:
- Language: { language }
- Tone: { tone }
- Length: { word_count | placeholder:Approximate word count | info: Suggested range: 500-2000 words }
Additional requirements:
- Include at least { source_count | type:enum | options:2, 3, 5 | label:Number of sources } credible sources
- Incorporate relevant statistics and data visualizations if applicable
- Address potential counterarguments or alternative viewpoints
Extra notes:
{ extra_notes | multiline | optional | placeholder: Any additional instructions or context }
Best Practices
- Be specific: Clearly define what you want the AI to do in your template.
- Use appropriate input types: Choose the right input type (text, multiline, enum) for each piece of information you need.
- Provide context: Use the
info attribute to give users helpful information about each input.
- Balance flexibility and structure: Offer enough customization options while maintaining a clear overall structure.
- Test your templates: Experiment with different combinations of inputs to ensure your template produces the desired results.
Conclusion
Prompt templates are a powerful tool for creating dynamic, customizable AI experiences. By mastering the use of tags, attributes, and built-in inputs, you can design sophisticated prompts that cater to a wide range of use cases and user needs. 
