parry-guard
github.com
Rust 编写的提示注入扫描器,通过本地 DeBERTa/Llama 模型检测 AI 编程工具中的注入攻击与密钥泄露
开源可自部署本地模型命令行LinuxMac
被推荐 1 次
适合做什么
适合使用 Claude Code、Codex 等 AI 编程工具的开发者,通过 hooks 拦截提示注入、密钥泄露和数据外泄
标签与属性
谁在推荐
Parry-guard by vaporif - Prompt injection scanner for AI coding tool hooks, written in Rust. Catches injection attacks, leaked secrets, and data exfiltration in tool inputs and outputs by running DeBERTa v3 and optionally Meta Llama Prompt Guard 2 locally via Candle. Built for Claude Code hooks; the models are gated on HuggingFace, so installation requires accepting their licences first. Self-desc
同类推荐
开源浏览器 Agent 框架,让 AI 像人一样操作网页,支持 Python/TypeScript 与云端浏览器。
开源可自部署支持 MCP网页
github.com1 处推荐