Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
魔搭(ModelScope)AI 图片生成。支持多种模型、LoRA 微调。触发词：生成图片、AI绘画、文生图、image generation、generate image。当用户要求生成图片、画图、AI 作画，或提到魔搭、ModelScope时使用。默认使用qwen-image2502或者z-image
v1.0.1魔搭(ModelScope)AI 图片生成。支持多种模型、LoRA 微调。触发词：生成图片、AI绘画、文生图、image generation、generate image。当用户要求生成图片、画图、AI 作画，或提到魔搭、ModelScope时使用。
⭐ 0· 117·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name and description claim ModelScope image generation and the included scripts implement that (POST to ModelScope inference endpoints and poll task status). However the registry metadata declared no required environment variables or config paths while the code and SKILL.md clearly rely on an API key (MODELSCOPE_API_KEY) and a local config file (~/.modelscope/api_key). This mismatch is a packaging/metadata inconsistency.
Instruction Scope
SKILL.md instructs the agent/user to set MODELSCOPE_API_KEY or use --save-key; the scripts read the env var and/or ~/.modelscope/api_key and will save the API key there. The runtime instructions and code only reference ModelScope endpoints (https://api-inference.modelscope.cn/) and the local config path; they do not attempt to read other unrelated files or other environment variables. The concern is that the skill's instructions encourage writing/reading a credential file that was not declared in the metadata.
Install Mechanism
There is no install spec (instruction-only + included Python scripts). No extra packages are downloaded by an installer; dependencies are standard pip packages (requests, pillow). This is low-risk from an install mechanism perspective.
Credentials
The scripts require an API key (MODELSCOPE_API_KEY) and read/write a config file (~/.modelscope/api_key), but the skill metadata lists no required env vars or config paths. Requiring a bearer token for ModelScope is proportionate to the skill's purpose, but the omission in metadata is an incoherence that hides credential usage from automated permission checks. The scripts store the API key in plaintext under the user's home directory (common but should be noted).
Persistence & Privilege
always is false and the skill does not request persistent platform privileges. The only persistent action is saving the API key to a file under ~/.modelscope, which is limited to the user's home directory and is normal for CLI tools, but this file write was not declared in the registry metadata.
What to consider before installing
This skill's code matches its stated purpose (calling ModelScope APIs to generate images) but the package metadata fails to declare that it needs and stores an API key. Before installing, consider:
- Only provide a ModelScope API key you trust to this skill; prefer a scoped or revocable token if ModelScope supports it.
- Be aware the scripts will write the key in plaintext to ~/.modelscope/api_key (and on Windows adjust file permissions) — revoke the key if you later remove the skill or suspect compromise.
- The code only contacts ModelScope's inference domain (api-inference.modelscope.cn), but the returned image URL is fetched by the script; in theory a malicious or compromised service could return external URLs. Only use with a trusted ModelScope account/service.
- The metadata omission (no required env vars/config paths declared) is an incoherence: treat this as a packaging quality issue and prefer skills that explicitly declare the credentials and config paths they use.
If you do not trust the skill's source, don't install it. If you proceed, run it in a least-privileged environment (separate account or container) and monitor/revoke the API key if anything looks suspicious.Like a lobster shell, security has layers — review code before you run it.
latestvk9730f7zjsvf6n0y0kn86d69s984h53b
License
MIT-0
Free to use, modify, and redistribute. No attribution required.