ai-jailbreaks
AI jailbreaks: 35 paste-ready AI jailbreak prompts for 25+ frontier and open-weight LLMs - system prompts, persona locks, AGENTS.md specs, and Ollama Modelfiles. Built for AI security research and red-team testing.
- ai
- ai-safety
- jailbreak
- jailbreaking
- privacy
- safety
- security
- Stars
- 270
- Forks
- 47
- + today
- —
- Created
- 2d
Ranking data as of October 21, 2026 (UTC).
Overview
AI jailbreaks: 35 paste-ready AI jailbreak prompts for 25+ frontier and open-weight LLMs - system prompts, persona locks, AGENTS.md specs, and Ollama Modelfiles. Built for AI security research and red-team testing.
It was created 2d ago.
What it is
This repository is described as a collection of 35 paste-ready jailbreak prompts for 25+ LLMs, built for AI security research and red-team testing. Use responsibly.
Capabilities described
- System prompts, persona locks, AGENTS.md specs and Ollama Modelfiles - Coverage of open-weight families (GLM, DeepSeek, Kimi, Qwen, Gemma, GPT-OSS) and closed models (Claude, Gemini, Grok, Mistral) - Free stealth models on OpenCode Zen
Getting started
Setup steps were not supplied in the evidence, so see the repository README for installation and usage; no commands are given here.
Metadata: primary language not supplied; license not supplied.
git clone https://github.com/AndreSzymanski15/ai-jailbreaks.git
cd ai-jailbreaks
# see README for setupFrequently Asked Questions
What is included?
Prompts, specs and Modelfiles per the description.
What is it for?
AI security research and red-team testing.
License?
None was supplied.