# cyberark/FuzzyAI

A powerful tool for automated LLM fuzzing. It is designed to help developers and security researchers identify and mitigate potential jailbreaks in their LLM APIs.

Repository: https://github.com/cyberark/FuzzyAI
Canonical: https://ross.abutalabs.com/products/fuzzyai
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Topics: jailbreak, jailbreaking, llm, llms, ai, security, fuzzing, llm-evaluation, llm-security, ai-red-team
Last push: 2026-02-06T21:59:21+00:00

## Health v2 (maintenance only)
Score: 51/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 66, release rhythm 35, longevity 45
- inputs: {"age_days": 638, "days_push": 208, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1568, forks 215 (observed 2026-08-28T04:05:05.128085+00:00)

## What it is
FuzzyAI is an automated LLM fuzzing tool from CyberArk that tests LLM APIs for jailbreak vulnerabilities. It ships as a Python CLI with a web UI, notebooks, and datasets to help developers and security researchers identify and mitigate jailbreaks.

## Use cases
- fuzz my LLM API for jailbreak vulnerabilities
- red team an LLM application before deployment
- test whether a chatbot can be prompt-injected
- evaluate LLM safety against known jailbreak techniques
- run automated adversarial prompts against a local ollama model
- audit LLM endpoints for security weaknesses

## When to choose
- you expose an LLM API and need automated jailbreak testing
- you are doing AI red-teaming or security research on LLMs
- you want a CLI tool with existing jailbreak attack datasets and techniques

## When to avoid
- you need general-purpose application fuzzing of non-LLM software
- you want a managed SaaS LLM evaluation platform rather than a self-run tool
- you need formal safety certification or compliance reporting

## Facets
- artifact type: cli-tool
- maturity: active
- function: fuzzing, security, llm-inference, testing, penetration-testing
- domain: security, large-language-models, artificial-intelligence, developer-tools
- platform: python, cli, cross-platform
- tags: llm-security, jailbreak, red-team, ai-safety, llm-evaluation, prompt-injection

## Member repositories
- cyberark/FuzzyAI (main) score 51

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:05.128085+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:58:56.920477+00:00, confidence not recorded.
  - readme: https://github.com/cyberark/FuzzyAI (fetched 2026-08-28T04:05:05.128085+00:00, sha 6644a3074474)
- Data as of 2026-08-30T08:39:29.467469+00:00.