← Back to Skills Marketplace
139
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install cogvideox-generator
Description
使用 CogVideoX (智谱AI) 模型根据文本提示词生成视频。
README (SKILL.md)
这是一个视频生成技能，调用智谱 AI 的 CogVideoX 接口。
参数说明
- prompt (string, 必填): 描述视频内容的提示词，越详细效果越好。
- quality (string, 可选): 输出模式，'quality' (质量优先) 或 'speed' (速度优先)。默认为 'quality'。
- size (string, 可选): 视频分辨率，如 '1920x1080'。默认为 '1920x1080'。
- fps (integer, 可选): 帧率，30 或 60。默认为 30。
- with_audio (boolean, 可选): 是否生成音频。默认为 True。
使用示例
用户说：“帮我生成一个比得兔开小汽车的视频。” AI 将调用此技能，参数为：{"prompt": "比得兔开小汽车，游走在马路上，表情开心"}
Usage Guidance
This skill appears to do what it says (generate video via ZhipuAI) and the script only uses one API key (ZHIPUAI_API_KEY). However: 1) the registry metadata incorrectly lists no required env vars while the code expects ZHIPUAI_API_KEY — confirm the platform will grant the skill that env var only if you trust it; 2) there is no install/dependency specification for the 'zai' SDK and no homepage or source repository for the skill or SDK — verify the provenance of the 'zai' package before installing; 3) since the source is unknown, consider running the skill in an isolated environment or reviewing the 'zai' library/package code and network endpoints it contacts; 4) only provide your ZhipuAI API key if you trust the skill owner; otherwise ask the publisher for a repo, release URL, and a dependency list (or a vetted install spec) before installing.
Capability Analysis
Type: OpenClaw Skill
Name: cogvideox-generator
Version: 1.0.2
The skill bundle is a standard implementation for generating videos using the ZhipuAI CogVideoX API. The script `scripts/generate_video.py` correctly handles the ZHIPUAI_API_KEY via environment variables and implements a straightforward asynchronous polling logic to retrieve video results. No indicators of malicious intent, data exfiltration, or prompt injection were found.
Capability Assessment
Purpose & Capability
The name/description, SKILL.md, and script all align: this is a CogVideoX (ZhipuAI) video-generation skill and it needs a ZHIPUAI_API_KEY. However registry metadata earlier reported "Required env vars: none" and the skill has no homepage/source URL, which is inconsistent and reduces transparency.
Instruction Scope
SKILL.md and the script only describe submitting a video-generation task and polling results from ZhipuAI. There are no instructions to read arbitrary local files, other env vars, or exfiltrate data to unexpected endpoints in the provided files.
Install Mechanism
There is no install spec even though the script imports 'zai' and uses ZhipuAiClient; the required Python dependency is not declared and the skill's source/homepage is missing. This is a usability/traceability gap and increases risk because fetched dependencies or missing libs could cause the integrator to install packages manually or run unclear code.
Credentials
The script only needs a single API key (ZHIPUAI_API_KEY), which is appropriate for calling the external service. But the registry metadata claiming no required env vars contradicts the code and SKILL.md, which is an inconsistency the user should note.
Persistence & Privilege
The skill does not request always:true, does not persist configuration or modify other skills, and does not require elevated or system-wide privileges.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install cogvideox-generator - After installation, invoke the skill by name or use
/cogvideox-generator - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Added project documentation file SKILL.md and project lock file .clawhub/lock.json.
- Removed legacy documentation file SKILLS.md.
- No changes to functionality; documentation and metadata updates only.
v1.0.1
- 添加了对 ZHIPUAI_API_KEY 权限的说明，需提供智谱AI的API密钥。
- 精简了技能简介，移除了音频、分辨率和帧率自定义的描述。
v1.0.0
---
name: cogvideox-generator
description: 使用 CogVideoX 模型根据文字描述生成短视频。
---
# CogVideoX 视频生成技能
该技能可以调用本地或远程部署的 CogVideoX 模型，将你的文字创意变为动态视频。
## 使用方法
向 AI 发送指令，例如：“请帮我生成一个‘一只猫在草地上追逐蝴蝶’的视频。”
Metadata
Frequently Asked Questions
What is cogvideox 生成视频?
使用 CogVideoX (智谱AI) 模型根据文本提示词生成视频。 It is an AI Agent Skill for Claude Code / OpenClaw, with 139 downloads so far.
How do I install cogvideox 生成视频?
Run "/install cogvideox-generator" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is cogvideox 生成视频 free?
Yes, cogvideox 生成视频 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does cogvideox 生成视频 support?
cogvideox 生成视频 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created cogvideox 生成视频?
It is built and maintained by King (@ilaus); the current version is v1.0.2.
More Skills