GhostWhisper AI: Bringing Ghost Stories to Life with Open-Source AI

GhostWhisper AI: The Offline Ghost Storyteller

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

What I Built

I built GhostWhisper AI, a spooky, interactive ghost story generator inspired by the theme "Ghost in the Machine."

The idea is simple: what if a machine could turn our memories, mysterious experiences, and imagination into eerie stories?

GhostWhisper AI lets users enter a story idea, select a horror genre, and generate a supernatural tale. It features a haunted-terminal interface, multiple horror themes, a story archive called the Memory Vault, and options to copy or download generated stories.

The project also supports an optional local AI mode using Ollama and an open-weight language model, allowing users to generate stories on their own computers instead of relying entirely on a cloud-based AI service.

Demo

Try the live website here:

GhostWhisper AI — Live Demo

Code

Explore the project on GitHub:

GhostWhisper AI — Source Code

The project uses HTML, CSS, and JavaScript to create the interface and manage story generation, local model requests, and the browser-based Memory Vault.

How I Built It

I built GhostWhisper AI using HTML, CSS, and JavaScript, focusing on a dark, atmospheric interface that feels like interacting with a mysterious machine.

The project has two storytelling modes:

• Demo Mode: Provides sample horror stories so users can explore the experience without installing an AI model.
• Local AI Mode: Connects to Ollama and can use an open-weight model such as Qwen2.5 3B to generate original stories locally.

The JavaScript handles user prompts, genre selection, story rendering, copying, downloading, and saving stories in browser local storage. The CSS creates the haunted-terminal aesthetic with glowing purple accents, subtle animations, and a responsive layout.

The local AI integration is the core open-source AI component. Once the model is downloaded and the local environment is configured, users can generate stories without sending their prompts to a third-party cloud AI API.

Why Does Open Innovation Matter?

For me, open innovation means having the freedom to build, experiment, and improve technology without being completely dependent on a single service provider.

GhostWhisper AI explores three important benefits:

• Privacy: Local inference can keep prompts on the user's computer instead of sending them to an external AI provider.
• Control and customization: Users can experiment with different compatible models and modify the prompts, interface, and storytelling behavior.
• Independence: Once the required software and model are installed, the local AI experience can operate without an internet connection, provided the local setup is configured correctly.

A closed API can be convenient, but it can introduce usage limits, ongoing costs, and dependence on an external service. An open-weight approach gives developers more control over how the AI runs and how the project evolves.

GhostWhisper AI is my experiment in making AI-powered creativity more personal, customizable, and independent.

My Agent Session

I did not include a DevRelay agent session in this submission. I focused on building and integrating the project myself.

The Spirit of It

The inspiration behind GhostWhisper AI is the idea that stories can keep memories alive, even when the people who created them are no longer around.

During Halloween and Día de los Muertos, stories about spirits, remembrance, and the connection between past and present feel especially meaningful. I wanted to bring that atmosphere into a small AI project.

The Memory Vault represents this idea: every saved story becomes something that can be revisited rather than forgotten.

The machine does not need to contain a real ghost. Sometimes, imagination, memory, and a little code are enough to make something feel alive.

Prize Categories

Open-source AI and local inference.