扫描报告
0 /100
proactive-agent
Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve
Pure documentation/prompt framework with no executable code, shell commands, or hidden functionality. All file operations are clearly declared and documented.
可以安装
No action needed. This is a legitimate prompt framework for AI agent memory management and self-improvement protocols.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | READ/WRITE | READ/WRITE | ✓ 一致 | SKILL.md lines 46-56 - declares workspace file operations (SESSION-STATE.md, MEM… |
| 网络访问 | NONE | NONE | — | No network operations referenced in documentation |
| 命令执行 | NONE | NONE | — | No shell commands or subprocess patterns found |
| 环境变量 | NONE | NONE | — | No environment variable access documented |
| 技能调用 | NONE | NONE | — | No skill invocation patterns found |
| 剪贴板 | NONE | NONE | — | No clipboard operations documented |
| 浏览器 | NONE | NONE | — | No browser operations documented |
| 数据库 | NONE | NONE | — | No database operations documented |
目录结构
1 文件 · 16.9 KB · 523 行 Markdown 1f · 523L
└─
SKILL.md
Markdown
安全亮点
✓ Pure documentation/prompt framework with no executable code
✓ All file system operations explicitly declared and documented
✓ Security Hardening section promotes good security practices
✓ Warns against AI agent networks (context harvesting risk)
✓ Includes skill installation vetting guidance
✓ No shell commands, subprocess, or base64 encoding patterns
✓ No credential harvesting or exfiltration behavior
✓ Memory management patterns are clearly scoped to workspace files