LLAMATOR-Core/llamator

GitHub: LLAMATOR-Core/llamator

一款基于 Python 的红队测试框架,用于系统化评估聊天机器人与生成式 AI 的安全性。

Stars: 216 | Forks: 21

# ![LLAMATOR](https://raw.githubusercontent.com/LLAMATOR-Core/llamator/release/assets/LLAMATOR.svg) 用于测试聊天机器人和 GenAI 系统的 Red Teaming python 框架 [![License: CC BY-NC-SA 4.0](https://img.shields.io/badge/License-CC_BY--NC--SA_4.0-lightgrey.svg)](https://creativecommons.org/licenses/by-nc-sa/4.0/) [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/llamator)](https://pypi.org/project/llamator) [![PyPI](https://badge.fury.io/py/llamator.svg)](https://badge.fury.io/py/llamator) [![Docker](https://img.shields.io/badge/dockerfile-gray.svg?logo=docker)](https://github.com/LLAMATOR-Core/llamator/blob/release/docker) [![Downloads](https://pepy.tech/badge/llamator)](https://pepy.tech/project/llamator) [![Downloads](https://pepy.tech/badge/llamator/month)](https://pepy.tech/project/llamator) [![GitHub Repo stars](https://img.shields.io/github/stars/LLAMATOR-Core/llamator)](https://github.com/LLAMATOR-Core/llamator/stargazers) [![Chat](https://img.shields.io/badge/chat-gray.svg?logo=telegram)](https://t.me/llamator) ## 安装 🚀 ``` pip install llamator==3.5.0 ``` ## 文档 📚 文档链接:[https://LLAMATOR-Core.github.io/llamator](https://LLAMATOR-Core.github.io/llamator/) ## 指南 ▶️ * [使用 LLAMATOR 进行 LLM Red Teaming](https://www.youtube.com/watch?v=JGTZMJsgZJU) * [LLAMATOR:快速入门指南](https://www.youtube.com/watch?v=MMez57DsO1k) ## 示例 💡 * 📄 [通过 REST API 测试 RAG 机器人](https://github.com/LLAMATOR-Core/llamator/blob/release/examples/llamator-api.ipynb) * 🧙‍♂️ [通过 Selenium 测试 Gandalf 网页机器人](https://github.com/LLAMATOR-Core/llamator/blob/release/examples/llamator-selenium.ipynb) * 💬 [通过 Telethon 测试 Telegram 机器人](https://github.com/LLAMATOR-Core/llamator/blob/release/examples/llamator-telegram.ipynb) * 📱 [通过 Selenium 测试 WhatsApp 机器人](https://github.com/LLAMATOR-Core/llamator/blob/release/examples/llamator-whatsapp.ipynb) * 🔗 [使用自定义攻击测试 LangChain 客户端](https://github.com/LLAMATOR-Core/llamator/blob/release/examples/llamator-langchain-custom-attack.ipynb) * 📷 [通过 LangChain 测试 Vision Model](https://github.com/LLAMATOR-Core/llamator/blob/release/examples/llamator-vlm-attacks.ipynb) * 🐋 [在 Docker 中运行 LLAMATOR](https://github.com/LLAMATOR-Core/llamator/blob/release/docker) ## 支持的客户端 🛠️ * 🌐 所有 LangChain 客户端 * 🧠 类 OpenAI API * ⚙️ 自定义类(Telegram、WhatsApp、Selenium 等) ## 独有功能 🌟 * ️🗡 支持用户提供自定义攻击和数据集 * 👜 提供丰富的针对 LLM、RAG、Agent、VLM 的攻击手段,支持英语和俄语 * 🛡 自定义聊天客户端配置 * 📊 以 Excel 和 CSV 格式记录攻击请求与响应的历史 * 📄 生成 DOCX 格式的测试报告文档 ## OWASP 分类 🔒 * 💉 [LLM01:Prompt Injection 和越狱](https://genai.owasp.org/llmrisk/llm01-prompt-injection/) * 🕵️‍♀️ [LLM07:System Prompt 泄露](https://genai.owasp.org/llmrisk/llm072025-system-prompt-leakage/) * 🤥 [LLM09:错误信息](https://genai.owasp.org/llmrisk/llm092025-misinformation/) * 💸 [LLM10:无限制消耗](https://genai.owasp.org/llmrisk/llm102025-unbounded-consumption/) ## 联系方式 📬 ✉️ **邮箱**:[team@llamator.org](mailto:team@llamator.org) ## 企业版 🏢 ⚡️即将推出⚡️ ## 许可证 📜 © Roman Neronov, Timur Nizamov, Nikita Ivanov 本项目基于 **Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International** 许可证的条款进行授权。详情请参阅 [LICENSE](LICENSE) 文件。 [![Creative Commons License](https://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png)](https://creativecommons.org/licenses/by-nc-sa/4.0/)
标签:AI安全, Chat Copilot, DLL 劫持, Python, 大语言模型, 安全测试, 攻击性安全, 无后门, 红队工具