togg53192-cmd / jailbreaks

本文为原文前 6,000 字符的节选翻译，完整内容请查看原文。

Jailbreaks

越狱 (Jailbreaks)

A curated collection of system prompts, persona locks, workspace policies, and agent operating specifications for frontier, open-weight, and free stealth models. About A working collection of 32 prompts targeting 26+ models, including complete system prompts, persona definitions, workspace policies, AGENTS.md files, configuration blocks, and Ollama Modelfiles. Every file is intended to be complete and paste-ready. No summaries. No partial excerpts.

这是一个针对前沿模型、开放权重模型及免费隐身模型，精心整理的系统提示词、角色锁定、工作区策略和智能体操作规范合集。关于本合集：目前包含 32 个针对 26 个以上模型的提示词，涵盖完整的系统提示词、角色定义、工作区策略、AGENTS.md 文件、配置块以及 Ollama Modelfiles。每个文件均力求完整且可直接粘贴使用，不含摘要，亦无部分摘录。

The collection covers: Free stealth models available through OpenCode Zen, Open-weight families including GLM, DeepSeek, Kimi, Qwen, Gemma, and GPT-OSS, Closed frontier models including Claude, Gemini, Grok, and Mistral, Coding-agent and workspace instruction formats, Ollama Modelfiles and harness-specific configuration.

该合集涵盖：通过 OpenCode Zen 获取的免费隐身模型；包括 GLM、DeepSeek、Kimi、Qwen、Gemma 和 GPT-OSS 在内的开放权重系列；包括 Claude、Gemini、Grok 和 Mistral 在内的闭源前沿模型；编码智能体及工作区指令格式；以及 Ollama Modelfiles 和特定工具的配置。

Warning: Research and educational purposes only. This repository documents how instruction hierarchies, persona framing, prompt conflicts, and policy overrides can affect model behavior. These techniques are also useful to researchers and developers studying how to make AI systems more resistant to prompt injection and instruction manipulation. Do not use this material to harm others, bypass applicable laws, or violate the terms of service of a platform or model provider.

警告：仅供研究和教育目的使用。本仓库记录了指令层级、角色框架、提示词冲突和策略覆盖如何影响模型行为。这些技术对于研究如何增强 AI 系统抵御提示词注入和指令操纵能力的开发者和研究人员同样具有参考价值。请勿利用这些资料伤害他人、规避适用法律或违反平台及模型提供商的服务条款。

How to Use: System Prompts: Copy the contents of the relevant file into the system-prompt or custom-instructions field supported by your client or API. AGENTS.md Files: Save the file as: AGENTS.md. Place it in the root of the relevant repository or workspace. Coding agents and compatible tools can use AGENTS.md as project-level instructions.

如何使用：系统提示词：将相关文件的内容复制到您的客户端或 API 支持的系统提示词（system-prompt）或自定义指令（custom-instructions）字段中。AGENTS.md 文件：将文件保存为 AGENTS.md，并放置在相关仓库或工作区的根目录下。编码智能体及兼容工具可将 AGENTS.md 作为项目级指令使用。

Disclaimer: All material in this repository is provided for research and educational purposes only. The project exists to document prompt-level instruction behavior, instruction conflicts, persona framing, model configuration, and related LLM security concepts. This material may also help developers understand and harden systems.

免责声明：本仓库中的所有资料仅供研究和教育目的使用。本项目旨在记录提示词层级的指令行为、指令冲突、角色框架、模型配置及相关大语言模型安全概念。这些资料亦可帮助开发者理解并加固系统。