← Back to Skills Marketplace
222
Downloads
0
Stars
3
Active Installs
1
Versions
Install in OpenClaw
/install modelscope-image
Description
通过魔搭社区(ModelScope) API 生成图片。先使用 --list-models 查看可用模型，然后根据用户需求由 AI 生成专业的提示词，最后调用 API 生成图片。支持 Kolors、Stable Diffusion XL、FLUX 等多种文生图模型。当用户需要使用魔搭社区、ModelScope 或...
Usage Guidance
This skill appears to do what it says: it generates prompts and calls ModelScope's image API using your MODELSCOPE_API_KEY. Before installing, ensure you trust the ModelScope service and limit the API key's permissions / billing exposure. Be aware the bundled model list includes models that may produce adult or sensitive content; avoid selecting models or prompts that violate policy or your organization's rules. Finally, the skill will make outbound network calls to api-inference.modelscope.cn (it needs the API key), so treat the key as sensitive and rotate/revoke it if you stop using the skill.
Capability Analysis
Type: OpenClaw Skill
Name: modelscope-image
Version: 1.0.0
The skill bundle is a legitimate tool for generating images using the ModelScope API. The core logic in `scripts/gen.py` uses Python's standard library (`urllib`) to communicate with official ModelScope endpoints (api-inference.modelscope.cn) and includes robust features like asynchronous task polling, model list parsing from `references/models.md`, and HTML gallery generation. No evidence of data exfiltration, malicious execution, or prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description, required binary (python3), primary env var (MODELSCOPE_API_KEY), and included script (scripts/gen.py) all align with a ModelScope image-generation helper. The requested resources are proportional to the stated functionality.
Instruction Scope
SKILL.md confines actions to reading local reference files, composing prompts, and calling ModelScope endpoints. It recommends launching a subagent to consult references/models.md (this is internal and token-saving). Note: the included model list contains entries that may enable generation of adult or sensitive content; prompt and model selection remain the user/agent responsibility.
Install Mechanism
Install spec only suggests installing Python via Homebrew (well-known package source). No arbitrary downloads or extract-from-URL steps are present.
Credentials
Only MODELSCOPE_API_KEY is required (declared as primary). The SKILL.md and code use only that key (or --api-key override). No unrelated credentials, secret-scoped paths, or broad environment access are requested.
Persistence & Privilege
always is false and the skill does not request elevated or persistent platform privileges. It does not modify other skills or global agent config in the provided files.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install modelscope-image - After installation, invoke the skill by name or use
/modelscope-image - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- modelscope-image-gen 功能的初始版本发布。
- 通过 ModelScope API 使用多种预配置模型（如 Kolors、Stable Diffusion XL、FLUX）实现图像生成。
- 提供工作流程：列出模型 → 选择模型（支持子代理） → AI 优化的提示生成 → 图像创建。
- 包含内置的提示工程指导和高质量结果场景，针对中文和多语言支持进行了优化。
- 命令行界面提供灵活的参数设置，可用于提示、模型、输出和模型列表。
- 文档提供详细的使用示例、排障指南以及其他资源的链接。
Metadata
Frequently Asked Questions
What is modelscope-image?
通过魔搭社区(ModelScope) API 生成图片。先使用 --list-models 查看可用模型，然后根据用户需求由 AI 生成专业的提示词，最后调用 API 生成图片。支持 Kolors、Stable Diffusion XL、FLUX 等多种文生图模型。当用户需要使用魔搭社区、ModelScope 或... It is an AI Agent Skill for Claude Code / OpenClaw, with 222 downloads so far.
How do I install modelscope-image?
Run "/install modelscope-image" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is modelscope-image free?
Yes, modelscope-image is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does modelscope-image support?
modelscope-image is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created modelscope-image?
It is built and maintained by baiyz0825 (@baiyz0825); the current version is v1.0.0.
More Skills