This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine
What I Built
BugSéance is a spooky AI debugging assistant that helps developers uncover the “ghost” hiding inside their code.
Developers can paste source code, an error message, or a stack trace and ask BugSéance to investigate. It responds with a ghostly personality while focusing on practical debugging help:
A summary of what went wrong
The likely root cause
Evidence from the submitted code or error
A suggested fix
Steps to verify the fix
Confidence and limitations when the cause is uncertain
I interpreted “Ghost in the Machine” as the mysterious, frustrating feeling developers get when a bug appears, disappears, or refuses to explain itself.
Instead of treating the ghost as just a visual theme, I made it the personality of a debugging companion — a spirit that helps developers investigate the unknown inside their code.
Code
👻 GitHub Repository : https://github.com/Gopika252006/BugSeance
The project includes:
React + Vite frontend
FastAPI backend
Ollama integration for local model inference
API endpoints for health, model availability, and bug analysis
A dark, haunted-terminal-inspired interface
How I Built It
BugSéance is built with an open-source-friendly stack:
Frontend: React, Vite, TypeScript, and CSS
Backend: Python and FastAPI
Local AI runtime: Ollama
Model: [Add the exact open-weight model you tested]
Testing: Pytest and frontend build checks; [add Playwright results if completed]
The frontend sends the developer’s code or error details to the FastAPI backend. The backend communicates with the configured Ollama instance and asks the model to produce a structured debugging analysis.
The model’s response is then displayed in the UI as a ghostly investigation report.
The goal is not to make the AI sound spooky at the expense of usefulness. The personality adds atmosphere, while the analysis stays focused on evidence, possible causes, and actionable fixes.
Why Does Open Innovation Matter?
Open innovation is central to BugSéance.
Using an open-weight model with local inference makes it possible to build a debugging assistant that developers can run and experiment with on their own machines.
This approach offers several important possibilities:
Privacy: Source code can remain on the developer’s machine when the application is configured for local inference.
Control: Developers can choose and replace the model based on their hardware and needs.
Customization: The prompts, model, and debugging workflow can be adapted by the community.
Accessibility: Developers can experiment without depending on a paid hosted AI API.
Offline potential: Once the runtime and model are installed, local inference can work without an internet connection.
Open innovation also means the project can be inspected, modified, improved, and extended by other developers instead of remaining a closed debugging service.
Top comments (0)