中文 EN

扫描报告

扫描新文件

⚠ 低风险 — 风险评分 15/100
上次扫描：22 小时前 重新扫描
15 /100
calling-agent-squad
Activate a multi-agent team (the Squad) to manage complex projects, business tasks, or development workflows
This is a legitimate multi-agent coordination framework with no malicious behavior detected. The skill coordinates role-based sub-agents for project management tasks through an orchestrator pattern.
技能名称calling-agent-squad
分析耗时40.8s
引擎pi
✓
可以安装
No immediate action required. Consider documenting allowed-tools requirements in SKILL.md for transparency. The hardcoded user path in squad-init.sh will cause failures on non-George systems.

安全发现 2 项

严重性 安全发现 位置
低危
Missing allowed-tools declaration 文档欺骗
SKILL.md does not specify required allowed-tools permissions, making it unclear what tools this skill needs.
No allowed-tools section in frontmatter
→ Add an allowed-tools declaration to SKILL.md frontmatter for transparency.
 SKILL.md:1
低危
Hardcoded user path reduces portability 供应链
squad-init.sh contains hardcoded path /Users/george/.openclaw which will fail on other systems.
BASE_DIR="/Users/george/.openclaw/workspace/skills/calling-agent-squad"
→ Use environment variables or a relative path for portability.
 squad-init.sh:3
资源类型声明权限推断权限状态证据
文件系统 READ READ ✓ 一致 Creates project folders in Documents/squad_projects/ per SKILL.md
命令执行 NONE READ ✓ 一致 squad-init.sh uses bash but is maintenance-only, not invoked during normal opera…
网络访问 NONE NONE — No network activity observed
环境变量 NONE NONE — No environment variable access detected
技能调用 NONE READ ✓ 一致 spawns sub-agents via openclaw agent (documented in SKILL.md)
剪贴板 NONE NONE — Not used
浏览器 NONE NONE — Not used
数据库 NONE NONE — Not used

目录结构

51 文件 · 86.2 KB · 2455 行
Markdown 50f · 2435L Shell 1f · 20L
├─ ▾ 📁 agents
│ ├─ ▾ 📁 architect
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 7L · 352 B
│ │ ├─ 📝 SOUL.md Markdown 6L · 566 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ ├─ ▾ 📁 brand-reviewer
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 8L · 424 B
│ │ ├─ 📝 SOUL.md Markdown 8L · 747 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ ├─ ▾ 📁 code-reviewer
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 8L · 394 B
│ │ ├─ 📝 SOUL.md Markdown 9L · 907 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ ├─ ▾ 📁 coder
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 8L · 368 B
│ │ ├─ 📝 SOUL.md Markdown 7L · 561 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ ├─ ▾ 📁 copywriter
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 7L · 286 B
│ │ ├─ 📝 SOUL.md Markdown 7L · 478 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ ├─ ▾ 📁 observer
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 8L · 385 B
│ │ ├─ 📝 SOUL.md Markdown 6L · 547 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ ├─ ▾ 📁 researcher
│ │ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ │ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ │ ├─ 📝 IDENTITY.md Markdown 8L · 308 B
│ │ ├─ 📝 SOUL.md Markdown 6L · 489 B
│ │ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ │ └─ 📝 USER.md Markdown 17L · 477 B
│ └─ ▾ 📁 squad-manager
│ ├─ 📝 AGENTS.md Markdown 212L · 7.7 KB
│ ├─ 📝 HEARTBEAT.md Markdown 5L · 168 B
│ ├─ 📝 IDENTITY.md Markdown 8L · 512 B
│ ├─ 📝 SOUL.md Markdown 15L · 1.2 KB
│ ├─ 📝 TOOLS.md Markdown 40L · 860 B
│ └─ 📝 USER.md Markdown 17L · 477 B
├─ ▾ 📁 templates
│ └─ 📝 handbook.md Markdown 13L · 348 B
├─ 📝 SKILL.md Markdown 104L · 3.5 KB
└─ 🔧 squad-init.sh Shell 20L · 758 B

安全亮点

✓ No credential harvesting - skill does not access ~/.ssh, ~/.aws, .env, or sensitive environment variables
✓ No data exfiltration - no external IP connections, POST requests, or data transmission observed
✓ No code obfuscation - no base64, eval(), or anti-analysis techniques found
✓ No remote script execution - no curl|bash, wget|sh, or similar patterns
✓ Behavior aligns with documentation - sub-agent spawning is declared in SKILL.md
✓ All 8 agent configs share identical AGENTS.md template - no hidden unique capabilities
✓ squad-init.sh only runs during explicit maintenance, not during normal skill operation