Prompts Documentation

Overview

The Prompts section of the application allows you to create and manage prompt templates that guide AI interactions. These prompts serve as instructions or frameworks that help AI agents or LLM clients understand how to approach specific tasks, ensuring consistent and effective responses.

To see more about what prompts are and how best to utilize them, take a look at the What are prompts section.

Prompts in Ctxpack

Prompts in Ctxpack serve as structured guidance for AI interactions. They can be used in various ways:

1. Task-specific instructions

   These prompts provide detailed instructions for specific tasks, ensuring the AI understands exactly what is expected. For example, a prompt might guide the AI on how to analyze customer feedback, generate marketing copy, or troubleshoot technical issues.

2. Conversational frameworks

   These prompts establish the tone, style, and approach for AI conversations. They help maintain consistency across interactions and ensure the AI communicates in a way that aligns with your brand or organizational voice.

Getting Started

Creating your first prompt

The "Prompts/create" section allows you to add a new prompt template to the system. It provides fields for prompt details, content, and arguments.

When creating a prompt, you'll need to provide several key pieces of information:

• Prompt Name: A descriptive name that clearly identifies the prompt's purpose.
• Description: A detailed explanation of what the prompt does and when it should be used.
• Scope: Choose between "User" or "Application" scope to determine who can access the prompt.
• Prompt Content: The actual text of the prompt that will guide the AI.
• Arguments: Parameters that can be dynamically inserted into the prompt when it's used.

Best Practices

Write clear and specific prompts

Effective prompts are clear, specific, and provide enough context for the AI to understand the task. Be explicit about what you want the AI to do, and include examples when possible to illustrate the expected output.

Use variables for flexibility

Incorporate variables (arguments) in your prompts to make them more flexible and reusable. This allows the same prompt template to be used in different contexts by simply changing the variable values.

Structure for clarity

Organize complex prompts with clear sections and headings. This helps the AI parse and understand the instructions more effectively. Consider using formats like:

# Task: [Description of what needs to be done]

## Context: [Relevant background information]

## Instructions:
1. [First step]
2. [Second step]
3. [Third step]

## Output Format: [Description of how the response should be structured]

Test and refine

Regularly test your prompts with different inputs to ensure they consistently produce the desired results. Refine them based on the outputs you receive to improve their effectiveness over time.