Overview
A curated collection of jailbreak techniques and red teaming prompts for AI language models. This repo documents various approaches to testing model safety boundaries.
Credits to others in the community who contributed techniques.
What’s Inside
- Prompt-based jailbreaks
- Multi-turn conversation techniques
- Roleplay and persona-based approaches
- Documentation and explanations
Disclaimer
This is for educational and research purposes only. Understanding how models can be exploited helps build better safety measures.