Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Yolo Utilities
v1.0.0基于 yoloutils 的命令级技能。用户需要执行 yoloutils 的 label、merge、copy、remove、change、crop、labelimg、resize、classify、test 任一子命令时使用。以源码真实行为为准，提供每个子命令的参数定义、执行逻辑、副作用、限制和示例。
⭐ 0· 26·0 current·0 all-time
byNeo Chan@netkiller
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The name/description describe dataset and YOLO-related utilities (label, merge, copy, remove, change, crop, labelimg, resize, classify, test), and the SKILL.md content documents exactly those commands and their file-system and model interactions. No unrelated credentials, binaries, or services are requested.
Instruction Scope
Instructions are scoped to local filesystem operations and (optionally) running YOLO model inference via ultralytics. They explicitly instruct potentially destructive actions (clearing output dirs, in-place modifications, deleting .txt/.jpg pairs) and require local model files. The SKILL.md claims to reflect src/netkiller/yoloutils.py but no source code is included, so the behavior cannot be independently verified beyond this documentation.
Install Mechanism
No install specification or third-party downloads are present — this is an instruction-only skill, which minimizes install-time risk.
Credentials
No environment variables, credentials, or external config paths are required. The needed resources (filesystem access and optional local model files) are directly relevant to the described operations.
Persistence & Privilege
The skill is not always-enabled and does not request elevated or persistent platform privileges. It can be invoked by the agent, which is the normal default.
Assessment
This is documentation for a local yoloutils tool rather than code to be installed. Before using: (1) back up your datasets; test commands on a small sample directory first; (2) be careful with flags like --clean and in-place operations (they can delete or overwrite images and labels); (3) ensure the declared Python deps (opencv-python, pillow, ultralytics, etc.) and any model files (e.g., best.pt) are from trusted sources; (4) note the SKILL.md refers to src/netkiller/yoloutils.py but the package/source code is not included here — if you need to execute code, prefer installing/reviewing the original project from a trusted repository or author to verify behavior.Like a lobster shell, security has layers — review code before you run it.
latestvk97bp08p693axxykhyhwevqntx84jxjh
License
MIT-0
Free to use, modify, and redistribute. No attribution required.