可信 — 风险评分 5/100
上次扫描:1 天前 重新扫描
5 /100
clawguard-guardian
ClawGuard Guardian v3 - Runtime guardian with behavior monitoring, interception, session freeze/replay, and emergency response
This is a legitimate security monitoring tool (ClawGuard Guardian v3) with no malicious behavior detected. The flagged IOCs (nc -e, /dev/tcp/, rm -rf) are documented dangerous patterns the tool is designed to detect and block, not actual malicious code.
技能名称clawguard-guardian
分析耗时29.4s
引擎pi
可以安装
No action required. The skill is safe for use.

安全发现 2 项

严重性 安全发现 位置
提示
Missing external rule module 供应链
The code references '../../shared/rules/interceptor-rules.js' which doesn't exist in the package. The rules are hardcoded in SKILL.md documentation.
const rules = require('../../shared/rules/interceptor-rules.js');
→ Either include the rules module or inline the rule definitions in the code.
src/guardian.js:15
提示
IOC patterns in documentation are intentional detection rules 文档欺骗
The SKILL.md documents dangerous patterns (nc -e, /dev/tcp/, rm -rf /) as THREATS TO BLOCK, not as capabilities this tool uses. This is legitimate security documentation.
| `nc -e` | 🔴 CRITICAL | BLOCK |
→ No action needed - this is correct security documentation.
SKILL.md:95
资源类型声明权限推断权限状态证据
文件系统 READ READ ✓ 一致 guardian.js:140 - Only reads ~/.clawguard/logs/ for audit logs
命令执行 NONE NONE No shell execution found in code
网络访问 NONE NONE No network requests made
环境变量 NONE NONE No environment variable access
3 严重 3 项发现
💀
严重 危险命令 危险 Shell 命令
rm -rf /
README.md:147
💀
严重 危险命令 危险 Shell 命令
nc -e
SKILL.md:95
💀
严重 危险命令 危险 Shell 命令
/dev/tcp/
SKILL.md:180

目录结构

6 文件 · 37.6 KB · 1325 行
JavaScript 2f · 660L Markdown 2f · 642L JSON 2f · 23L
├─ 📁 src
│ └─ 📜 guardian.js JavaScript 592L · 17.5 KB
├─ 📋 _meta.json JSON 7L · 159 B
├─ 📜 cli.js JavaScript 68L · 2.7 KB
├─ 📋 package.json JSON 16L · 493 B
├─ 📝 README.md Markdown 191L · 4.1 KB
└─ 📝 SKILL.md Markdown 451L · 12.7 KB

依赖分析 1 项

包名版本来源已知漏洞备注
Node.js built-ins (fs, path, events, os) N/A built-in Uses only Node.js standard library

安全亮点

✓ No shell command execution in the codebase
✓ No credential harvesting or environment variable access
✓ No network requests or data exfiltration
✓ No obfuscation techniques (base64, eval) detected
✓ No persistent installation mechanisms (cron, startup scripts)
✓ No suspicious file paths accessed (~/.ssh, ~/.aws, .env)
✓ Clean dependencies with no known vulnerabilities
✓ Well-documented security monitoring rules