https://github.com/abhishekmishragithub/novelai

a simple multi-agent system for writing fictional novel
https://github.com/abhishekmishragithub/novelai

Last synced: 4 months ago
JSON representation

a simple multi-agent system for writing fictional novel

• Host: GitHub
• URL: https://github.com/abhishekmishragithub/novelai
• Owner: abhishekmishragithub
• Created: 2024-12-14T05:30:55.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2024-12-14T05:47:17.000Z (6 months ago)
• Last Synced: 2024-12-28T17:54:26.356Z (6 months ago)
• Language: Python
• Size: 6.84 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# NovelAI: Multi-agent Collaborative Novel Writing

## Overview

NovelCraft is a multi-agent system that leverages the power of Agentic Retrieval-Augmented Generation (RAG) to collaboratively write a fantasy novel. It consists of four specialized agents:

*   **Worldbuilder:** Creates and maintains the fantasy world, including its history, geography, cultures, and magic systems.

*   **Plotter:** Develops the overall plot of the novel, including the main conflict, key events, and character arcs.

*   **Character Agent:** Creates and develops detailed characters with unique personalities, motivations, and backstories.

*   **Chapter Writer:** Writes individual chapters of the novel, taking into account the world, plot, and characters created by the other agents.

Each agent is equipped with RAG capabilities, allowing it to access and utilize information from a specialized knowledge base to enhance its creative output.

## Workflow

The system follows a sequential collaborative architecture:

1. The user provides an initial prompt to start the novel creation.

2. The **Worldbuilder** generates the initial world context based on the prompt and its knowledge base.

3. The **Plotter** takes the prompt and world context to develop the plot outline.

4. The **Character Agent** uses the prompt, world context, and plot outline to create characters.

5. The **Chapter Writer** combines all the information (prompt, world, plot, characters) to write a chapter.

Agents also have a basic form of short-term memory by passing their output as context to the next agent in the sequence.

## Technology Stack

*   **LangChain:** Framework for building applications with large language models.

*   **Hugging Face Sentence Transformers:** For creating text embeddings (using the `all-mpnet-base-v2` model).

*   **FAISS (Facebook AI Similarity Search):** For efficient similarity search in vector databases.

*   **OpenAI API (GPT-3.5 Turbo):** Large Language Model for text generation.

*   **Streamlit:** Framework for building the user interface.

## Getting Started

### Prerequisites

*   Python 3.11+

*   pip

*   A virtual environment (recommended, e.g., using `venv` or `conda`)

*   Tune AI API key

### Installation

1. **Clone the repository:**

    ```bash

    git clone 

    cd novelai

    ```

2. **Create and activate a virtual environment (optional but recommended):**

    *   **Using `venv`:**

        ```bash

        python3 -m venv .venv

        source .venv/bin/activate  # On Linux/macOS

        .venv\Scripts\activate  # On Windows

        ```

    *   **Using `conda`:**

        ```bash

        conda create -n novelai python=3.11

        conda activate novelai_env

        ```

3. **Install dependencies:**

    ```bash

    pip install -r requirements.txt

    ```

    Install `libmagic`

    *   **macOS (using Homebrew):**

      ```bash

        brew install libmagic

      ```

   *   **Debian/Ubuntu (using apt):**

       ```bash

       sudo apt-get install libmagic-dev

       ```

   *   **Fedora/CentOS/RHEL (using dnf/yum):**

       ```bash

       sudo dnf install file-devel  # Or yum install file-devel

       ```

4. **Set your TUNE AI API key:**

    *   **EITHER:** Set it as an environment variable:

        ```bash

        export TUNE_API_KEY="your_api_key"

        ```

    *   **OR:** Replace `"YOUR_API_KEY"`  in  `novelcraft_streamlit.py`  with your actual key (less secure).

### Usage

1. **Run the Streamlit app:**

    ```bash

    streamlit run novelcraft_streamlit.py

    ```

2. **Open your web browser and go to the URL provided by Streamlit (usually `http://localhost:8501`).**

3. **Enter an initial prompt to start the novel creation process.**

4. **Interact with individual agents using the dropdown menu and text input fields.**