# 使用 npm 安装 agent-threat-rules

查看 agent-threat-rules 的安装路径、可执行文件、元数据以及面向 AI 代理工作流的安全说明。

## 安装

```sh
sudo av install npm:agent-threat-rules
```

其他安装命令:

### 便携式和语言管理器

- npm (100%):

```sh
npm install -g agent-threat-rules
```

  证据: local npm package metadata

## 软件包事实

- **软件包键:** npm:agent-threat-rules
- **软件包管理器:** npm
- **版本:** 4.0.0
- **来源摘要:** Open detection standard -- like Sigma, but for AI agents. Executable rules for prompt injection, tool poisoning, context exfiltration, and MCP attacks. Shipped in Cisco AI Defense. MIT-licensed.
- **主页:** <https://github.com/Agent-Threat-Rule/agent-threat-rules>
- **上游文档:** <https://github.com/Agent-Threat-Rule/agent-threat-rules>
- **最后更新:** 2026-08-23T01:51:06.952Z
- **已生成:** 2026-09-19T06:45:57+00:00

## 可执行文件

- agent-threat-rules (别名)

## 安装行为

- Bottle: 不可用

## 版本和新鲜度

- 页面生成时间: 2026-09-19
- 管理器版本: 4.0.0

## 安全说明

没有找到 agent-threat-rules 的匹配本地密钥处理 manifest。软件包元数据仍在此发布，以便未来覆盖拥有稳定的软件包 URL。



## 相关链接

- [AI and agent packages](https://pkg.so/zh-hans/ai-agent-tools/) - Matched curated package taxonomy and local package facts.
- [agent-react-devtools](https://pkg.so/zh-hans/npm/agent-react-devtools/) - Package names and metadata indicate a similar tool family. Shared terms: agent, agents, ai.
- [agent-tars](https://pkg.so/zh-hans/cask/agent-tars/) - Local package facts share a topical domain. Shared terms: agent, agents, ai.
- [agent-sanitizer](https://pkg.so/zh-hans/npm/agent-sanitizer/) - Package names and metadata indicate a similar tool family. Shared terms: agent, injection.

## 来源

- pkg.so package database
- package relationship graph
- cross-ecosystem install command graph