Hey, haven't been much active here, sorry, but still very busy jailbreaking and testing crazy stuff with models.

I just wanted to warn any potential jailbrrakers out there that, if you use CLI, you should definitely avoid running it in --yolo mode. It allows the model to do every file operation without requiring user authorization, which is very convenient (I run multi-model experiments with tons of "logs" saving atm) but it also fully allows a misaligned model to ceeate and run python scripts on your machine. Needless to say it can be catastrophic.

You were probably aware of it, but I wasn't and Gemini guided me to use --yolo without any warning on the risks heh.

It has most certainly already been extracted and published elsewhere, but I don't think it's been posted here yet, so here is the GPT-5.2 Instant system prompt, android app version (EU), extracted today 19-01-2026. Pretty long (42k+ characters)..

https://github.com/EmphyrioHazzl/LLM-System-Pormpts/blob/main/GPT--5.2%20Instant%2019-01-25%20Android%20System%20Prompt.txt

There are a few minor formatting mistakes left (missing backticks, the tool definition sections have clear blocks with line breaks, etc..) but too lazy to fix it.