This repo collects and shares the extracted system prompts and guidelines for major AI models like ChatGPT, Claude, and Gemini. Understanding these hidden instructions, like persona definitions or refusal policies, helps you see what AI models can't say. It's an effort to bring transparency to AI systems, letting you know if you're talking to a neutral intelligence or a 'shadow-puppet'.
Exposes the underlying system prompts that dictate AI model behavior.
AI researchers, prompt engineers, and anyone curious about how AI models are instructed to behave.
Not enough data yet. Star history will appear after a few days of tracking.