Phantom Shell — a haunted terminal possessed by a local open-weight model

Description and use

Phantom Shell is a local AI shell that comments on commands and supports chat without cloud services, ensuring privacy and offline operation.

Problem

Enable a local AI model to provide contextual commentary and chat within a terminal without relying on cloud APIs, preserving privacy and offline capability.

Outcome

Phantom Shell provides an on‑device AI shell experience with optional mock mode, enabling privacy‑centric, offline usage and easy model swapping.

Workflow

• Clone the repository: git clone https://github.com/pr7426/ghost-in-the-machine
• Navigate into the repo: cd ghost-in-the-machine
• Run demo with built‑in mock ghost: python3 -m phantom_shell demo or ./scripts/demo.sh
• Start interactive possess mode: python3 -m phantom_shell possess
• Optionally pull a real model: ollama pull llama3.2 then run python3 -m phantom_shell possess --model llama3.2

Evidence

Phantom Shell is a CLI that lets an open-weight model possess your terminal.

Source

If Ollama is not installed, a built-in mock ghost still demos the full UX so judges (and you) can try it offline in seconds.

Source

Theme take: the “ghost” should live in the machine — open weights, local inference, no cloud API key. Privacy is the haunting.

Source