📄 中文摘要
随着 GPT-4 或 Claude 的集成，LLM 应用程序的用户体验不断提升。然而，提示注入攻击可能导致系统泄露敏感信息。为了解决这一问题，开发了 AI Guardian 库，旨在防范此类安全隐患。该库提供了一系列工具，帮助开发者识别和防止提示注入攻击。通过实践，开发者可以更好地理解 LLM 的安全性及其潜在风险，从而提升应用程序的安全性和用户信任。
如何在 Python 中构建开源 LLM 安全库（以及我对提示注入的认识）
📄 English Summary
How I Built an Open-Source LLM Security Library in Python (and What I Learned About Prompt Injection)
Integrating GPT-4 or Claude into applications enhances user experience, but prompt injection attacks pose significant risks, potentially exposing sensitive information. To address this issue, the AI Guardian library was developed to mitigate such security vulnerabilities. This library offers a set of tools that assist developers in identifying and preventing prompt injection attacks. Through practical implementation, developers can gain a deeper understanding of LLM security and its potential risks, thereby enhancing the safety and user trust of their applications.