扫描报告
5 /100
mouse-yolo-factory
Mouse YOLO Factory - YOLO integration skill for mouse product defect detection including scratch generation, auto-labeling, and dataset merging
This is a legitimate YOLO-based computer vision tool for mouse product defect detection with no malicious behavior detected.
可以安装
This skill is safe to use. No security concerns were identified.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | NONE | READ | ✓ 一致 | All scripts perform image I/O operations |
| 文件系统 | NONE | WRITE | ✓ 一致 | Creates output directories and saves processed images |
| 网络访问 | NONE | READ | ✓ 一致 | YOLO model loading is standard ML behavior |
| 命令执行 | NONE | NONE | — | No shell execution found |
| 环境变量 | NONE | NONE | — | No environment variable access |
| 凭证 | NONE | NONE | — | No credential access detected |
目录结构
5 文件 · 45.9 KB · 1250 行 Python 4f · 1219L
Markdown 1f · 31L
├─
DatasetManager.py
Python
├─
datatool.py
Python
├─
drawbox_and_dataset_savejson_with_model.py
Python
├─
Mouse_produce_scratch.py
Python
└─
SKILL.md
Markdown
依赖分析 6 项
| 包名 | 版本 | 来源 | 已知漏洞 | 备注 |
|---|---|---|---|---|
ultralytics | * | pip | 否 | YOLO framework - standard ML library |
opencv-python | * | pip | 否 | Standard computer vision library |
numpy | * | pip | 否 | Standard numerical library |
torch | * | pip | 否 | PyTorch - standard ML framework |
tqdm | * | pip | 否 | Progress bar utility |
pandas | * | pip | 否 | Data manipulation library |
安全亮点
✓ Uses standard, well-known ML libraries (ultralytics, opencv, numpy, torch)
✓ No network exfiltration or C2 communication
✓ No credential harvesting or environment variable iteration
✓ No obfuscation techniques (base64, eval, etc.)
✓ Documentation accurately describes functionality
✓ No remote script execution or curl|bash patterns
✓ No access to sensitive paths (~/.ssh, ~/.aws, .env)
✓ No supply chain risks detected in dependencies