pjcampbe11 Goto Github PK

adopting-adversarial-approaches---llms

adversarial-attacks

ai-audio-startups

Community list of startups working with AI in audio and music technology

ai-exploits

A collection of real world AI/ML exploits for responsibly disclosed vulnerabilities

art

🎨 ASCII art library for Python

attack-kubernetes

Ref to Docker Images Developed for Kubernetes Security Assessments

attacking-llms_with-llms

Red Teaming LLMs with LLMs A Guide to LLM Security Assessments

autodan

The official implementation of our ICLR2024 paper "AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models".

awesome-ai-music-generation

A curated compilation of AI-driven generative music resources and projects. Explore the blend of machine learning algorithms and musical creativity.

awesome-backdoor-in-deep-learning

A curated list of papers & resources on backdoor attacks and defenses in deep learning.

awesome-llm-cybersecurity-tools

A curated list of large language model tools for cybersecurity research.

awesome-ml-security

backdoor-learning-resources

A list of backdoor learning resources

charcuterie

Data Scientists Go To Jupyter

chat-models-backdoor-attacking

Code for the paper "Exploring Backdoor Vulnerabilities of Chat Models"

chatgpt-prompts-bug-bounty-refined

ChatGPT Prompts for Bug Bounty & Pentesting

codellama

Inference code for CodeLlama models

datasets

🤗 The largest hub of ready-to-use datasets for ML models with fast, easy-to-use and efficient data manipulation tools

easyjailbreak

An easy-to-use Python framework to generate adversarial jailbreak prompts.

evals

Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.

evasivegpt

fickling

A Python pickling decompiler and static analyzer

garak

LLM vulnerability scanner - https://github.com/leondz/garak/fork

gotorefs

hero

a hackbot proof-of-concept

hip

The Hide Injected Prompts (HIP) script is designed to provide a wide range of functionalities through a command-line interface. Each command is associated with a specific operation, allowing users to perform complex text processing, encryption, encoding, and other tasks with ease.

inject-prompts

jailbreak-gemini-1.5-pro-api-0514

jailbreaks

jinja

A very fast and expressive template engine.