Skip to content
TFD-42 edited this page Aug 12, 2026 · 2 revisions

FAQ

Do I need Ollama for this tool? No. --no-ollama (CLI) or the "Ollama en secours" checkbox left unticked (GUI) runs the full classify + enrich pipeline with pattern/keyword matching alone and zero network calls. Ollama is only a fallback for files keyword matching can't confidently classify or scan for placeholders.

Is this tool legal to use? Yes, but only for authorized security testing and research. Always obtain proper authorization before using any payload against a system you don't own or don't have explicit permission to test. See ETHICS.md.

What script formats are supported? Ducky Script: .txt, .duck, .ds.

Can I use this commercially? Yes, under the MIT license. Please include license attribution — see LICENSE.

Does this tool claim authorship of the payloads it organizes? No. It classifies and organizes third-party community payloads sourced from the repos in Bad_USB_Classifier/url.txt. Full credit and copyright remain with each original author — see Source Repositories & Credits.

CLI or GUI — which should I use? Either produces the same result. The CLI (badusb_pipeline.py) is faster for scripting/automation; the GUI (badusb-gui) is better if you'd rather drag & drop and fill in a form than remember flags.

Where does my Discord webhook / attacker IP get stored? Nowhere but the output scripts you generate yourself, on your own machine. The enrichment agent writes values directly into the copied scripts in your chosen output folder — nothing is sent anywhere by the tool itself.

Does this work without an internet connection? The core classify + enrich pipeline does, with --no-ollama. Fetching new source repos (--urls, discover_repos.py) and the optional Ollama fallback obviously need network access.

Clone this wiki locally