The code in this repository is proudly and entirely written using AI

PlotKing - AI Story Builder

PlotKing is an interactive AI Book Generator that helps writers generate, validate, refine, and export novels. The application takes a short plot description and transforms it into a complete book using local or cloud LLM deployments.

🚀 How It Works (General Flow)

PlotKing follows a structured pipeline to ensure high-quality output:

1. Refine: You define the core idea, genre, and structure. You can refine this manually, use AI refine to automatically expand and enhance your plot, or have a conversation with Plot King AI to brainstorm and detail your story concept.
2. Create: Using the refined input from the previous step, the AI expands this into a detailed plot, chapter overview, and full chapters.
3. Edit: You review the generated content. You can rewrite sections with AI, chat with the AI for suggestions, or manually edit text. The unique Validation System ensures your changes don't break the story's continuity.
4. Export: Once satisfied, you export your novel to a formatted EPUB file, ready for reading or publishing.

---

✨ Key Features

1. Refine: Shape Your Story Concept

Before creating your book, refine your story idea through multiple approaches:

• Manual Input: Directly write and edit your plot description, genre, and structure.
• AI Refine: Use the AI refine button to automatically expand and enhance your plot with creative details, proper structure, and genre-appropriate elements.
• Chat with Plot King: Have a conversation with Plot King AI to brainstorm ideas, discuss characters, settings, conflicts, and narrative details. The AI helps you explore different aspects of your story and fills in gaps.

2. Create: From Idea to Structure

The creation process uses your refined input and is broken down into logical steps to maintain coherence:

• Plot Expansion: Turn your refined idea into a rich, multi-paragraph synopsis.
• Chapter Outline: The system generates an outline of chapters based on the expanded plot.
• Full Generation: Each chapter is written individually, with the AI referencing the plot and previous chapters to maintain context.
• Automatic Validation: After generating a chapter, a separate AI agent reads it to ensure it matches the outline. If it fails, it auto-regenerates until it passes.

3. Edit: Advanced AI-Assisted Writing

The Editor tab is where the magic happens. It's not just a text box; it's a suite of tools:

🛠️ Editing Modes

• View: Read-only mode to browse your story.
• Manual: Standard text editor for direct changes.
• Rewrite: Select text and give instructions (e.g., "Make this dialogue more intense", "Describe the setting in more detail").
• Chat: Converse with "Plot King" to brainstorm ideas or request changes for the current section.

🛡️ Validation System & Drafts

When you make changes (manually or via AI), PlotKing doesn't just overwrite your work. It creates Drafts.

• Impact Analysis: The system analyzes how your change affects future chapters.
• Auto-Generated Drafts: If your change alters the plot significantly, the AI suggests updates for subsequent chapters to maintain continuity.
• Draft Review: You can compare the Checkpoint (original) vs. Draft (new) using the comparison view ⚖️.
• Selective Apply: You choose which AI suggestions to keep and which to discard or regenerate.

4. Export: Publish Ready

Turn your project into a finished product:

• EPUB Generation: Creates a professional e-book file compatible with most readers.
• Metadata Fetching: AI can suggest a title based on your story content.
• Customization: Choose fonts (Georgia, Palatino, etc.) and set font sizes for the output file.
• Cover Art: Upload a custom cover image or generate one using AI, to be embedded in the EPUB.

5. Project Management

Keep your stories organized:

• Save: Persist your current progress, including plot and chapters.
• Load: Switch between different story projects easily.
• New: Start a fresh project from scratch.
• Delete: Remove unwanted projects.

6. Settings: Configurable Models & Task Assignments

• Models: Support for multiple providers (LM Studio, OpenAI, Gemini, xAI for LLM tasks; Automatic1111 and OpenAI for image generation) with configurable local endpoints, API keys, and provider-specific settings.
• Tasks: Assign different models to specific tasks for optimal performance and easy switching.

---

Requirements

The application requires:

• Python 3.6+
• Gradio for the UI
• Access to an LLM API (default: http://localhost:1234/v1/chat/completions for local deployments, but configurable)
• Additional dependencies listed in requirements.txt

Getting Started

1. Install dependencies: pip install -r requirements.txt
2. Run the application: python main.py or ./run.bat on Windows.
3. Access the UI through your browser at http://localhost:7860

Running a Local LLM with LM Studio

PlotKing is designed to work seamlessly with local LLM deployments, and LM Studio provides an easy way to run models locally without internet dependency.

🔹 Step 1: Install LM Studio

1. Download and install LM Studio for your operating system: 👉 https://lmstudio.ai/
2. Launch LM Studio after installation.

🔹 Step 2: Load a Local Model

1. Open the Models tab in LM Studio.
2. Search for and download a model suitable for long context and creative writing. For local development, I'm using gpt-oss 20b.
3. Once downloaded, load the model in memory.

💡 Tip: For long-form book generation, prefer models that support at least 128K context windows.

🔹 Step 3: Enable the Local Server API

1. In LM Studio, open the Developer tab (top-right gear icon).
2. Toggle "Start Server".
3. The default address is http://127.0.0.1:1234/v1/chat/completions (this works out of the box), but you can configure a different endpoint in PlotKing's Settings tab if needed.

🔹 Step 4: Enable FlashAttention (Optional, for Faster Generation)

If your GPU supports it (NVIDIA RTX 30xx, 40xx, or newer), you can enable FlashAttention for significant performance improvements, especially when working with long context windows.

In LM Studio while the model is loaded:

1. Open the model's Settings, the gear on the left of the loaded model.
2. Enable “FlashAttention”
3. Restart the model to apply changes.

Running a Local Image Generator with Automatic1111

PlotKing supports Stable Diffusion for generating book covers via the Automatic1111 WebUI.

🔹 Step 1: Install Automatic1111

Follow the installation instructions for your OS: 👉 https://github.com/AUTOMATIC1111/stable-diffusion-webui

🔹 Step 2: Enable the API

To allow PlotKing to communicate with Stable Diffusion, you must enable the API and set the port.

1. Open the webui-user.bat file (Windows) or webui-user.sh (Linux/Mac) in a text editor.
2. Add the following arguments to the COMMANDLINE_ARGS variable (port 6969 is suggested to match PlotKing's default configuration, but you can use any port and configure it in PlotKing's Settings tab):

   set COMMANDLINE_ARGS=--api --port 6969

3. Save the file and run it to start the server.

Contributing

We would love your help building PlotKing! To make collaboration effective, please adhere to the following:

• Keep Pull Requests Focused: Each Pull Request should address a single issue or add one specific feature. Please do not mix bug fixes, new features, and code refactoring in the same PR.
• Target the develop Branch: All Pull Requests must be opened against the dev branch. PRs opened against the master branch will be closed.

Future Work

PlotKing is an evolving project. Upcoming features, experimental ideas, and long-term development milestones are documented in the Roadmap.

Check it out to see what’s planned next — from interactive editing tools and story infill/outfill features to multi-model support and visual story generation.