Prompt-Framework 1.0.0

Prompt_Framework is a Python package that provides a set of flexible frameworks for prompt engineering. It allows seamless interchangability between various frameworks

Prompt Framework

Prompt_Framework is a Python package that provides a set of flexible frameworks for prompt engineering. It allows seamless interchangability between various frameworks such as RACE, CARE, APE, CREATE, TAG, CREO, RISE, PAIN, COAST, ROSES, and REACT to build sophisticated prompts for language models with different context and task-based structures.

Link: https://github.com/Subhagatoadak/Prompt_Framework

---

Features

• Multiple Framework Support: Includes popular prompt engineering frameworks like RACE, CARE, APE, CREATE, TAG, CREO, RISE, PAIN, COAST, ROSES, and REACT.
• Seamless Switching: Easily switch between frameworks without writing new prompt logic.
• Customizable Prompts: Add common parameters during initialization and use framework-specific arguments as needed.
• Flexible Input: Accepts various parameters, allowing customization of prompts for different use cases.

---

Installation

To install Prompt_Framework, follow these steps:

1. Install via pip (for local development)

Clone the repository and install the package in editable mode:

git clone https://github.com/yourusername/Prompt_Framework.git
cd Prompt_Framework
pip install -e .

Usage

Once you’ve installed the package, you can easily import and start using the Prompt_Framework in your Python scripts.

1. Import the PromptFramework Class

First, import the PromptFramework class into your Python script:

from Prompt_Framework import Prompt_Framework

# Initialize with common parameters
prompt_tool = Prompt_Framework(context="Customer service inquiry.", output_type="solution", style="polite")

2. Switch to a Framework

To generate a prompt for a specific framework, use the switch_framework() method. For instance, you can switch to the RACE framework:

# Switch to the RACE framework and generate a prompt
prompt_tool.switch_framework("race")
race_prompt = prompt_tool.generate_prompt(
    role="Assistant",
    action="Provide customer service assistance",
    explanation="Provide a detailed and polite response based on the inquiry."
)

print(race_prompt)

Switch Between Frameworks

You can easily switch between different frameworks as per your needs. For example, after generating a RACE prompt, you might want to switch to the APE framework:

# Switch to APE framework and generate a prompt
prompt_tool.switch_framework("ape")
ape_prompt = prompt_tool.generate_prompt(
    action="Solve account issue",
    purpose="Help the user recover their account",
    execution="Provide detailed account recovery instructions."
)

print(ape_prompt)

Customizing Frameworks

Frameworks come with specific requirements for the prompt parameters. For example:

RACE Framework

The RACE framework might require:

• role: The role of the assistant (e.g., "Assistant").
• action: What action the assistant should take (e.g., "Provide customer service assistance").
• explanation: The explanation of the task (e.g., "Provide a detailed and polite response based on the inquiry.").

# Customizing RACE framework
prompt_tool.switch_framework("race")
race_prompt = prompt_tool.generate_prompt(
    role="Customer Service Representative",
    action="Assist with product inquiry",
    explanation="Provide helpful and polite product details in response to customer queries."
)

print(race_prompt)

Frameworks Included

The following prompt engineering frameworks are included in the package:

1. RACE

Role: Specify role Action: Mention action needed Context: Provide background information Explanation: Describe the outcome

2. CARE

Context: Provide background information Action: Mention action needed Result: State the goal Example: Give example outputs

3. APE

Action: Define the job to be done Purpose: State the goal Execution: Describe the desired outcome

4. CREATE

Character: Specify the role Request: Define the job to be done Examples: Provide example outputs Adjustment: Suggestions for improvement Type of Output: Define the output format Extras: Additional context

5. TAG

Task: Define the task Action: Define job to be done Goal: Describe the end goal

6. CREO

Context: Provide background information Request: Define the job to be done Explanation: Explain the task Outcome: Describe the desired outcome

7. RISE

Role: Mention the role Input: Provide context and instructions Steps: Ask for step-by-step instructions Execution: Describe the desired outcome

8. PAIN

Problem: Describe the problem Action: Mention the action needed Information: Request any necessary details Next Steps: Ask for resources or next steps

9. COAST

Context: Provide background information Objective: Define the goal Actions: List actions required Scenario: Describe the scenario Task: Define the task to be completed

10. ROSES

Role: Define the role of the responder Objective: State the expected result Scenario: Provide background context Expected Solution: Describe the expected solution Steps: Ask for the steps to achieve the solution

11. REACT

Context: Provide background information Task: Define the task to be completed Explanation: Describe the task or problem

API Reference

PromptFramework Class Initialization

prompt_tool = Prompt_Framework(context="Context information", output_type="desired_output", style="desired_style")

context (str): Background information for the prompt. output_type (str): The expected output format (e.g., "solution", "steps"). style (str): Style of the output (e.g., "polite", "casual"). switch_framework(framework_name)

Switches to a specific framework.

framework_name (str): The name of the framework to switch to. Options include "race", "ape", "create", etc.

generate_prompt(*args, **kwargs)

Generates a prompt using the currently selected framework.

args: Framework-specific arguments. kwargs: Additional keyword arguments for customization.

Contributing

We welcome contributions! If you'd like to improve this package, feel free to fork the repository, make your changes, and submit a pull request. Please make sure to write tests for new features or bug fixes.

Steps to Contribute:

Fork the repository Clone your fork: git clone https://github.com/yourusername/Prompt_Framework.git Create a new branch: git checkout -b feature-branch Make your changes Push your changes: git push origin feature-branch Create a pull request

License

This project is licensed under the MIT License - see the LICENSE file for details.