← Back to Skills Marketplace
1017
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install tianshu-wan-video
Description
使用通义万相 2.6 生成视频，支持文生视频和图生视频。Node.js 实现，无需 Python。
README (SKILL.md)
通义万相视频 (tianshu-wan-video)
直接调用阿里云通义万相 2.6 视频生成模型，Node.js 实现。
功能
- 文生视频 (t2v)
- 图生视频 (i2v)
前置配置
DASHSCOPE_API_KEY- 阿里云 DashScope API Key
用法
# 文生视频
node scripts/generate_video.js t2v --prompt "电影感特写镜头，缓慢推进" --duration 5 --resolution 720P
# 图生视频
node scripts/generate_video.js i2v --prompt "镜头缓慢旋转" --image-url "https://example.com/image.jpg" --duration 4
输出
脚本输出 VIDEO_URL: \x3Curl>，提取该 URL 即可使用。
Usage Guidance
This skill appears to do what it says: it runs a Node.js script that calls Alibaba's DashScope video API and needs only DASHSCOPE_API_KEY. Before installing, consider: (1) the package has no published homepage or repository and the owner is unknown—this reduces trustworthiness; (2) the script makes network calls to dashscope.aliyuncs.com, so the API key will be sent to that service (expected for use); (3) ensure you only provide a DashScope API key you trust the skill to use. If you need higher assurance, ask the author for a code repository or verify the script locally in a controlled environment (check Node version compatibility and run it manually) before granting the key to an agent that could call it autonomously.
Capability Analysis
Type: OpenClaw Skill
Name: tianshu-wan-video
Version: 1.0.0
The skill bundle is a legitimate implementation for generating videos using Alibaba's DashScope (Tongyi Wanxiang) API. The core logic in `scripts/generate_video.js` uses standard fetch requests to official endpoints, handles API keys appropriately via environment variables, and contains no evidence of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
The name/description say it generates videos with 通义万相 (DashScope) and the script calls https://dashscope.aliyuncs.com/api/v1 with a bearer token. Requiring DASHSCOPE_API_KEY is consistent and expected.
Instruction Scope
SKILL.md instructs running the included Node.js script with prompt/image-url args. The script only reads the declared DASHSCOPE_API_KEY env var and CLI args, calls the DashScope API, polls for task completion, and prints VIDEO_URL. It does not access unrelated files, other env vars, or external endpoints beyond the DashScope domain.
Install Mechanism
No install spec (instruction-only + small script). package.json has no dependencies; the script relies on global fetch (Node 18+). No downloads or archive extraction are present.
Credentials
Only one required credential (DASHSCOPE_API_KEY) is declared and used for Authorization in API calls. No other secrets or unrelated environment variables are requested.
Persistence & Privilege
The skill does not request always:true or any elevated persistent privileges, nor does it modify other skills or system configs. Autonomous invocation is allowed by default but not combined with other red flags.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tianshu-wan-video - After installation, invoke the skill by name or use
/tianshu-wan-video - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Node.js 实现，替代 qwen-wan-video
Metadata
Frequently Asked Questions
What is Tianshu Wan Video?
使用通义万相 2.6 生成视频，支持文生视频和图生视频。Node.js 实现，无需 Python。 It is an AI Agent Skill for Claude Code / OpenClaw, with 1017 downloads so far.
How do I install Tianshu Wan Video?
Run "/install tianshu-wan-video" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Tianshu Wan Video free?
Yes, Tianshu Wan Video is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Tianshu Wan Video support?
Tianshu Wan Video is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Tianshu Wan Video?
It is built and maintained by tianshu (@wangshengli0421); the current version is v1.0.0.
More Skills