There are 9 repositories under prompt-injection topic.
Self-hardening firewall for large language models
Every practical and proposed defense against prompt injection.
Prompts of GPT-4V & DALL-E3 to full utilize the multi-modal ability. GPT4V Prompts, DALL-E3 Prompts.
ChatGPT Jailbreaks, GPT Assistants Prompt Leaks, GPTs Prompt Injection, LLM Prompt Security, Super Prompts, Prompt Hack, Prompt Security, Ai Prompt Engineering, Adversarial Machine Learning.
prompt attack-defense, prompt Injection, reverse engineering notes and examples | 提示词对抗、破解例子与笔记
Prompt injection attacks and defenses in LLM-integrated applications
A benchmark for prompt injection detection systems.
A prompt injection game to collect data for robust ML research
My inputs for the LLM Gandalf made by Lakera
Website Prompt Injection is a concept that allows for the injection of prompts into an AI system via a website's. This technique exploits the interaction between users, websites, and AI systems to execute specific prompts that influence AI behavior.
Build production ready apps for GPT using Node.js & TypeScript
The Prompt Injection Testing Tool is a Python script designed to assess the security of your AI system's prompt handling against a predefined list of user prompts commonly used for injection attacks. This tool utilizes the OpenAI GPT-3.5 model to generate responses to system-user prompt pairs and outputs the results to a CSV file for analysis.
Whispers in the Machine: Confidentiality in LLM-integrated Systems
Turning Gandalf against itself. Use LLMs to automate playing Lakera Gandalf challenge without needing to set up an account with a platform provider.
Image Prompt Injection is a Python script that demonstrates how to embed a secret prompt within an image using steganography techniques. This hidden prompt can be later extracted by an AI system for analysis, enabling covert communication with AI models through images.
ASCII Art Prompt Injection is a novel approach to hacking AI assistants using ASCII art. This project leverages the distracting nature of ASCII art to bypass security measures and inject prompts into large language models, such as GPT-4, leading them to provide unintended or harmful responses.
Guard your LangChain applications against prompt injection with Lakera ChainGuard.
The Security Toolkit for managing Generative AI(especially LLMs) and Supervised Learning processes(Learning and Inference).
This project investigates the security of large language models by performing binary classification of a set of input prompts to discover malicious prompts. Several approaches have been analyzed using classical ML algorithms, a trained LLM model, and a fine-tuned LLM model.
AI/LLM Prompt Injection List is a curated collection of prompts designed for testing AI or Large Language Models (LLMs) for prompt injection vulnerabilities. This list aims to provide a comprehensive set of prompts that can be used to evaluate the behavior of AI or LLM systems when exposed to different types of inputs.
A prompt defence is a multi-layer defence that can be used to protect your applications against prompt injection attacks.
MER is a software that identifies and highlights manipulative communication in text from human conversations and AI-generated responses. MER benchmarks language models for manipulative expressions, fostering development of transparency and safety in AI. It also supports manipulation victims by detecting manipulative patterns in human communication.
This project leverages the SDXL-Turbo model for versatile image processing tasks. Offering a simple command-line interface, it facilitates both Text-to-image and Image-to-image operations. Users select an operation, input prompts, and the script dynamically generates and executes code snippets.
Let Raccoon sample the unknown, safeguarding your AI's home.
A serverless set of functions for evaluating whether incoming messages to an LLM system seem to contain instances of prompt injection; uses cascading cosine similarity and ROUGLE-L calculation against known good and bad prompts
Client SDK to send LLM interactions to Vibranium Dome
This repo focus on how to deal with prompt injection problem faced by LLMs