GhostTrace AI: A Local AI That Investigates Digital Ghosts

Hacktoberfest Open-Source AI Challenge: Ghost in the Machine Submission 👻🌼

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

👻 GhostTrace AI is a local AI-powered paranormal investigation storyteller that transforms fictional computer activity logs into eerie, detective-style ghost stories.

Inspired by Halloween and the mysterious events that might be hidden inside machines, I wanted to build a project that combines spooky storytelling with practical AI engineering.

GhostTrace AI analyzes sample activity logs, identifies unusual-looking events using transparent rule-based scoring, and uses a locally running language model to turn those events into a story.

GhostTrace AI is designed for fictional investigations and entertainment. It does not claim to detect real ghosts or supernatural activity.

See GhostTrace AI investigate mysterious sample events, display anomaly scores, and turn digital clues into a spooky story.

The repository contains the Python application, sample logs, automated tests, setup instructions, and project documentation.

I built GhostTrace AI using Python, Ollama, Qwen2.5 3B, Streamlit, and Pytest.

The application has five main components:

I also incorporated safety measures such as read-only log processing, path validation, file-size limits, and graceful handling of malformed input.

The project includes automated tests for log parsing, anomaly scoring, sample cases, timeline filtering, and story-generation fallback. During development, the reported test suite passed all 25 tests.

Open innovation is central to GhostTrace AI because it lets me build an AI-powered experience without depending on a paid, hosted AI API.

By running the open-weight Qwen2.5 3B model locally through Ollama, the application can process its investigation data on the user's machine. There are no per-request cloud AI charges, and developers can inspect the source code, modify prompts, and experiment with other supported models.

Local inference also provides greater control over data privacy. Once the required model and dependencies are installed, the application can continue using its fallback engine without an internet connection.

There are trade-offs, too. Local inference depends on the user's hardware, and the model must be downloaded before offline use. Story generation can also be slower on a CPU than on more powerful hardware.

For me, open innovation means having the freedom to experiment, customize, and build AI experiences without locking the entire project into a closed service.

I used Antigravity to help develop and test GhostTrace AI.

The project includes automated tests, sample investigation cases, and documentation describing the implementation.

Imagine opening an investigation and finding a process that wakes up at 3:33 AM, a forgotten server that appears active, or a file that seems to return after disappearing.

The mystery detector identifies the clues, explains why they were flagged, and the local AI turns them into a narrative.

The events are fictional, but the experience demonstrates how AI can combine structured data with creative storytelling.

GhostTrace AI is my small experiment in making open-source AI a little spooky—and showing that even a machine's logs can inspire a mystery.

This project focuses on the Ghost in the Machine theme and local open-source AI. I will list any applicable partner prize categories after confirming the official eligibility requirements.

Built with Python, local AI, and a little digital mystery. 👻💻