GitHub repository for a tool that detects and filters malicious prompts before they are entered into a Retrieval-Augmented Generation (RAG) database, ensuring data integrity and security.
security
research
jailbreak
prompt
gpt
malicious
database-security
zero-click
llm
prompt-engineering
generative-ai
prompt-injection
retrieval-augmented-generation
ai-worm
-
Updated
Jul 6, 2024 - Jupyter Notebook