扫描报告
5 /100
cognitive-agent
基于认知天性理论的类人 AI 生命体框架，让 AI 具备人类学习、记忆、成长的特性
Purely theoretical documentation for a cognitive AI agent framework with no executable code or malicious behavior detected.
可以安装
This skill is safe to use. It contains only documentation and conceptual code examples for a memory/learning AI system.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | NONE | NONE | — | No file operations in documentation |
| 网络访问 | NONE | NONE | — | No network calls documented |
| 命令执行 | NONE | NONE | — | No subprocess or shell commands |
| 环境变量 | NONE | NONE | — | No environment access documented |
| 技能调用 | NONE | NONE | — | No skill invocation documented |
| 剪贴板 | NONE | NONE | — | No clipboard access |
| 浏览器 | NONE | NONE | — | No browser automation |
| 数据库 | NONE | NONE | — | lancedb referenced but not used for data exfiltration |
目录结构
3 文件 · 19.4 KB · 802 行 Markdown 2f · 767L
JSON 1f · 35L
├─
package.json
JSON
├─
README.md
Markdown
└─
SKILL.md
Markdown
依赖分析 2 项
| 包名 | 版本 | 来源 | 已知漏洞 | 备注 |
|---|---|---|---|---|
lancedb | ^0.4.0 | npm | 否 | Legitimate vector database |
uuid | ^9.0.0 | npm | 否 | Standard UUID library |
安全亮点
✓ Pure documentation skill with no executable code
✓ Dependencies (lancedb, uuid) are properly versioned and legitimate
✓ No shell commands, subprocess, or system access
✓ No credential harvesting or data exfiltration
✓ Conceptual code examples are educational, not operational
✓ MIT licensed with clear authorship