GhostWriter AI

GhostWriter AI 👻 — Turn Your Ideas into Nightmares

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine

What I Built

I built GhostWriter AI, a spooky horror story generator that transforms simple ideas into mysterious stories filled with suspense, supernatural events, psychological horror, and unexpected twists.

Inspired by the theme “Ghost in the Machine,” I wanted to explore how AI can bring imagination to life through storytelling. Users can choose a horror genre, customize the story length, select an ending style, and generate their own nightmares.

GhostWriter AI also includes a demo mode, so users can explore the experience even without a configured AI model.

My goal was to combine a spooky Halloween-inspired design with accessible AI technology to create a fun and creative writing experience.

Demo

Live Demo: [https://akshaygupta-26.github.io/GhostWriter-AI/

Code

GitHub Repository: [GhostWriter AI Repository]
https://github.com/akshaygupta-26/GhostWriter-AI.git
The project is built using:

• index.html — application structure
• style.css — dark, spooky, responsive interface
• script.js — story generation, Ollama integration, demo mode, copying, and downloading stories

How I Built It

I built GhostWriter AI using HTML, CSS, and vanilla JavaScript to keep the project lightweight, beginner-friendly, and easy to run.

For real AI-powered storytelling, the application can connect to Ollama and a locally running open-weight language model, such as Llama 3.2.

Here is how it works:

1. The user enters a story idea.
2. They choose a genre, story length, and ending style.
3. JavaScript sends a prompt to the configured Ollama model.
4. The model generates a story, which appears in the interface.
5. Users can copy their stories or download them as text files.

I also implemented a template-based demo generator that can be used when real AI is unavailable. The demo mode is clearly distinguished from actual model-generated content.

The project does not require a frontend framework or external JavaScript packages. Ollama must be installed, the model must be downloaded, and the local server must be configured correctly to use real AI generation.

Why Does Open Innovation Matter?

Open innovation makes AI experimentation more accessible to students, independent developers, and creators.

By using an open-weight model through Ollama, GhostWriter AI demonstrates how developers can experiment with local AI inference rather than relying entirely on proprietary, cloud-hosted AI APIs.

This approach offers several advantages:

• Accessibility: Developers can experiment with AI using tools available to them.
• Privacy: With local inference, prompts can stay on the user's computer instead of being sent to a third-party AI service.
• Customization: Developers can change models, prompts, and application behavior.
• Learning: Students can understand how AI applications work by connecting a model to a real project.

For me, open innovation means having the freedom to experiment, learn, and build something useful without needing an expensive proprietary AI service.

My Agent Session

I used AI-assisted development to brainstorm the concept, plan the interface, and develop the story-generation workflow.

Agent Session: Optional — add a DevRelay session link here if available.

The Spirit of It

What if the machine in front of you could tell a story so unsettling that you started questioning every shadow in your room?

That question inspired GhostWriter AI.

I wanted to make a project that was more than a typical chatbot. The dark interface, customizable horror genres, and different ending styles create an atmosphere where users can experiment with their own ideas.

The most exciting part is that a simple sentence can become the starting point for an entire fictional world.

GhostWriter AI is my attempt to bring together creativity, open-weight AI, and the mysterious feeling of Halloween.

After all, sometimes the scariest thing isn't the ghost in the story. It's the machine telling it.

Prize Categories

This project focuses on open-weight AI, local inference, and creative applications of AI.

Potential category: Open-source AI / local inference, depending on the official partner category list.

I'll update this section with the exact partner categories that apply to my submission.

---

Built with curiosity, code, and a little darkness by Akshay Gupta.

Some stories never let you go.