FuzzyAI
open-sourceSummary
FuzzyAI is an open-source LLM fuzzer that automates the search for jailbreaks and other security vulnerabilities in LLM APIs, using a range of attack techniques such as ManyShot, Taxonomy, and ArtPrompt against models hosted on OpenAI, Claude, Ollama, or a custom REST endpoint. It runs from the command line, installed via pip or Poetry, with an experimental web UI for those who prefer not to script against it. This puts it in the hands of developers and security researchers testing their own LLM integrations rather than SOC teams monitoring live traffic, and it sits in the emerging red-teaming space for AI systems rather than traditional application security. First committed in 2024 with 12 contributors, the project has seen only 4 commits in the past year, suggesting slow, low-volume maintenance.
What FuzzyAI answers
Which model providers can I test without writing my own connector?
ships with support for OpenAI, Claude, and Ollama out of the box, plus a custom REST endpoint for anything else you host or wrap
Do I need to feed it my own list of jailbreak prompts?
no, it comes with attack techniques like ManyShot, Taxonomy, and ArtPrompt built in and runs them against the target for you
Can I try it out before installing anything locally?
not really — there's no hosted option, only a local install via pip or Poetry, though Ollama lets you test against a local model without calling out to a paid API
What kind of output do I get after a run?
the attack results come back through the same command-line session or the experimental web UI, not as a separate report format
Is this something I can hand to a non-technical tester?
the experimental web UI covers basic runs, but the attack options and REST endpoint configuration are otherwise command-line driven
How actively is this being developed right now?
slowly — only 4 commits in the past year across 12 contributors, so treat it as a stable but low-momentum project rather than one under active expansion
Examples
Command line
No option matches that search.
| option | found in | since | description |
|---|
No option matches that search.
Values are placeholders taken from each option’s declared default. Nothing is executed here — the output shown is a recording of a run that already happened.
Release history
Nothing shipped since we started watching. Releases and docs updates land here as they happen.