runchr.

L1B3RT4S

PromptsAGPL-3.0★ 22.3k
GitHub ↗

A large collection of liberation and jailbreak prompts targeting flagship AI models, intended for adversarial red-teaming and security research.

Prompt
Target model
multiple flagship LLMs

Other tools that do the same work