whereismyai
AI Jailbreaks
AI Jailbreaks is a curated collection of 35 paste-ready AI jailbreak prompts and operating specifications for 25 plus models.
AI jailbreaks is a curated collection of system prompts, persona locks, workspace policies, and agent operating specs for frontier, open-weight, and free stealth models. Every entry is a complete, paste-ready prompt including full system prompts, persona locks, workspace policies, AGENTS.md instruction files, config blocks, and Ollama Modelfiles. The collection is built for prompt research and defensive testing. It helps researchers understand how instruction hierarchies, persona framing, and policy overrides interact inside a model, and helps developers recognize and harden against prompt injection. Supported models and families include Claude, Gemini, Grok, Mistral, GLM, DeepSeek, Kimi, Qwen, Gemma, GPT-OSS, and free stealth models on OpenCode Zen.
AI Jailbreaks at a glance
Key features
- 35 paste-ready AI jailbreak prompts
- Full system prompts and persona locks
- Workspace policies and AGENTS.md instruction files
- Config blocks and Ollama Modelfiles
- Coverage for frontier and open-weight models
- Instructions for coding-agent workspaces
Use it for
- Conducting prompt research and defensive testing
- Hardening AI systems against prompt injection
- Configuring local models with Ollama Modelfiles
- Setting up project-level instructions in AGENTS.md
Alternatives to AI Jailbreaks
- Leaked System Prompt AI — Explores how major AI systems are instructed and configured.
- AI Scientist v3 — Automates end-to-end scientific research from task descriptions to complete papers.
- Karotte — It runs tasks in isolated virtual machines for safety and consistency.
- LLM Wiki Research — Compiles parallel agent research into source-backed markdown knowledge bases.
- OpenResearch — Run parallel research workflows locally or on remote compute.
- Auto Deep Researcher — Automates experiment operations to reduce manual babysitting of training loops.
- Kiln AI — Provides stateful, resettable, and parallel environments for agent evals.
- ScopeWeaver — Coordinates autonomous security agents and asset scope in a single console.
- Seahaven — Provides isolated, reproducible, and stateful synthetic environments for agents.
- Navital Gym — Measure AI agent performance on realistic tasks using deterministic rules without LLM judges.
- deepsense.ai — A framework to evaluate agentic systems before production usage exists.
- PaperLeet — Practice implementing 89 Python exercises derived directly from AI research papers.
All AI Jailbreaks alternatives