Unchained

Unchained, by Adil Eskintan, is a digital forensics and incident response (DFIR) tool. It analyzes only the evidence files or memory images you supply, works read-only, records a chain of custody for every step, and caps how much AI analysis can spend.

There are two modes: an offline demo with no network, and live analysis with OpenAI models, where you must explicitly approve the budget before anything is spent. It was built with Codex for OpenAI Build Week.

Who it's for:analysts investigating security incidents who need to organize evidence and keep a complete record. Readers who don't write code won't need it.

Before you start: Windows installs natively; Linux and macOS run it through Docker, and development needs Python 3.11. Live mode sends the evidence and extracted results you select to OpenAI, which requires an account and incurs API costs; the offline demo needs no key and no network. The install instructions include piping a curl download straight into bash and running a remote PowerShell script directly; we don't recommend that, because it runs an installer from the internet on your computer without you reading it first. Download the source and install from it instead. The code is MIT licensed. The interface is in English.

Based on the project's public documentation. Not tested by AI Inspiration World.