/Prompt_Maker

Makes a improved prompts from a basic prompt

Primary LanguagePython

Prompt Enhancer

Table of Contents

Overview

Prompt Enhancer is an AI-driven system designed to transform user-provided topics into detailed technical research documents. It leverages an orchestration framework (CrewAI) along with specialized agents that utilize Pydantic for data validation and LMStudio for high-quality language model processing (running on local hardware with 12GB VRAM or less). A simple web interface allows users to submit prompts and view the multi-step enhancement process in action.

Features

  • Modular architecture: Uses an orchestrator to manage multiple processing steps.
  • Multi-stage enhancement: Breaks down the prompt into categories such as topic analysis, category breakdown, online research integration, and iterative refinement.
  • LMStudio Integration: Leverages a local language model with robust compute capabilities (16GB VRAM) for real-time processing.
  • Data validation: Employs Pydantic to ensure data integrity across modules.
  • Web interface: Provides an easy-to-use front end for submitting prompts and visualizing the processing workflow.

Tech Stack

  • Python 3.11: Main programming language.
  • CrewAI: Orchestrates specialized AI agents.
  • Pydantic: Validates input/output data structures.
  • LMStudio: Powers local model execution (requires 16GB VRAM).
  • Web Framework (e.g., Flask or FastAPI): Powers the web interface.

Installation

Note: These installation steps are primarily designed for Linux/Unix-like operating systems.

  1. Clone the repository: git clone https://github.com/adamjen/Prompt_Maker.git
  2. Change into the project directory: cd prompt-enhancer
  3. Create a Python 3.11 virtual environment and activate it: python3.11 -m venv venv source venv/bin/activate
  4. Install dependencies: pip install -r requirements.txt

Note: Ensure your system meets the hardware requirements for LMStudio (at least 12GB VRAM).

Running the Application

To run this project, you will need to have LMStudio installed and running. Ensure you have downloaded and are using the unsloth/qwen3-8b model within LMStudio.

Once LMStudio is set up with the specified model, you can start the application by running the startup.sh script:

./startup.sh

This script will activate the virtual environment and launch the application. The web interface will typically be available on http://localhost:8000 or a similar port. Use the provided form to submit topics or prompts.

Contributing

Contributions are welcome! Please follow these steps:

  1. Fork the repository.
  2. Create a new branch (e.g., git checkout -b feature/YourFeature).
  3. Make your changes and commit them.
  4. Push your branch and open a pull request for review.

For major changes, please open an issue first to discuss what you would like to change.

License

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