Jailbreaks is an open-source reference collection of system prompts, persona configurations, and operating instructions for various AI language models. It documents techniques for influencing model behavior through instruction hierarchy manipulation and persona framing. The repository includes complete prompt templates for Claude, Gemini, Grok, DeepSeek, Qwen, GLM, and other model families. Its materials cover coding agents, workspace rules, assistant personalities, and model-specific configuration formats. Developers can study how conflicting instructions and policy overrides affect model responses. The collection also includes AGENTS.md configurations and Ollama Modelfiles for compatible environments. It is intended for educational research into prompt behavior, instruction robustness, and AI security testing rather than an executable software application.

Features

  • Curated AI System Prompt Collection
  • Model-Specific Persona and Behavior Configurations
  • Instruction Hierarchy and Prompt Conflict Examples
  • AGENTS.md Workspace Policy Templates
  • Ollama Modelfile and Coding Agent Configurations
  • Prompt Injection and AI Robustness Research

Project Samples

Jailbreaks Screenshot 1

Project Activity

See All Activity >

Categories

Libraries

Follow Jailbreaks

Jailbreaks Web Site

Other Useful Business Software
$300 Free Credits to Build on Google Cloud Icon
$300 Free Credits to Build on Google Cloud

New customers can spin up VMs, build with AI, and query data at no cost.

Put your $300 in credit toward real workloads, then keep building with free monthly usage for 20+ products. No commitment and no charge until you upgrade.
Start Free
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of Jailbreaks!

Additional Project Details

Registered

2 days ago
Report inappropriate content