cdn-url-transfer
将模型配置中的外部示例 URL(fal.media、googleapis 等)转存到速推 CDN(cdn-video.51sux.com),并更新代码和校验 API 返回。当用户提到"转存 CDN"、"替换示例链接"、"转存链接"、"CDN 地址"、"示例 URL 替换"时使用此 skill。
Why use this skill?
Automate the transfer of external AI model media URLs to your custom CDN. Ensure stable resource access and standardized file management with this OpenClaw skill.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/hexiaochun/cdn-url-transferWhat This Skill Does
The cdn-url-transfer skill is a specialized utility designed to stabilize model resource management within the OpenClaw ecosystem. Many AI models rely on external media URLs (e.g., fal.media, googleapis) that are prone to latency, expiration, or intermittent accessibility issues. This skill automates the process of migrating these assets to our high-performance CDN (cdn-video.51sux.com). By centralizing storage, the skill ensures that model examples and input parameters remain consistently accessible. The process involves identifying external URLs within model_registry.py and executor files, downloading them securely using local curl commands to bypass proxy restrictions, and uploading them to the TOS (Tencent Object Storage) bucket using a standardized naming convention.
Installation
You can install this skill directly via the OpenClaw hub CLI:
clawhub install openclaw/skills/skills/hexiaochun/cdn-url-transfer
Use Cases
- Production Stability: Migrating external demo content for public-facing AI models to ensure 100% uptime for documentation.
- Asset Standardization: Converting messy, externally hosted media URLs into a structured naming format within our custom CDN architecture.
- Deployment Readiness: Preparing model codebases for release by sanitizing references to third-party file hosters.
- Troubleshooting Broken Examples: When user reports indicate that model output examples or input templates are failing to load due to 403 or 404 errors on external source servers.
Example Prompts
- "转存 CDN: 请将 omnihuman 模型 registry 中的所有外部示例链接迁移到速推 CDN,并更新对应的 Python 文件。"
- "帮我替换示例链接,这些 fal.media 的图片源不稳定,请按照规范转存到我们的 bucket 中。"
- "检测到模型输出报错,可能是因为外链失效,请执行 CDN 地址转存任务。"
Tips & Limitations
- Base64 Constraint: While the skill supports
upload_image, use it only for files smaller than 100KB. For all other media (video/audio/large images), strictly use the TOS Python SDK transmission method to prevent memory overflows or network timeouts. - Naming Conventions: Always adhere to the
{模型简称}-{examples}/folder structure. This is critical for automated auditing and future cleanup tasks. - Cleanup: If you use the catbox temporary storage method, ensure you delete the local temporary file in
/tmp/immediately after a successful TOS transfer to maintain environment cleanliness. - Content Type: Always explicitly define the
content_type(e.g., video/mp4, image/png) when using the TOS SDK to ensure browsers render the assets correctly.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-hexiaochun-cdn-url-transfer": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, external-api, code-execution
Related Skills
水浒传故事小人书
水浒传故事小人书创建。使用 Nano Banana Pro 模型生成手绘卡通风格的水浒传故事信息图。当用户想要创建水浒传故事插画、小人书、信息图时使用此 skill。
style-extractor
从参考剧本或参考素材中提取统一风格锚点(STYLE_BASE),确保全剧视觉一致性。当需要匹配参考风格、提取画风、建立风格基准、生成风格资产包时使用。
视频链接解析
解析视频分享链接,获取无水印视频下载地址。当用户想要下载视频、解析抖音/快手/小红书/B站链接、获取无水印视频时使用此 skill。
vidu-video
使用 Vidu Q3 Pro 模型生成视频。当用户想要文生视频、生成带音频的视频,或提到 vidu 时使用此 skill。
character-creator
创建AI角色的完整流程,包括生成详细角色描述、文生图肖像和多角度参考图。使用即梦4.5模型。当用户要求创建角色、生成人物立绘、制作角色参考图、或需要多角度人物图时使用此技能。