Skill转LoomLoom工作流
将 OpenClaw Skill（SKILL.md）转化为 LoomLoom TemplateSpec v2 JSON。支持：单步/多步文本生成、图片生成、视频生成、文件上传输入、多模态输入、并行+汇总。自动识别 skill 结构并生成对应的工作流模板。不支持依赖工具调用（exec/read/write）的 skill。
更新于 2026年9月4日版本 v10输出类型 text
审核通过
输入与最终输出
输入
你需要提供
| 输入字段 | 类型 | 要求 | 说明 |
|---|---|---|---|
| SKILL.md 内容 | 文本 | 必填 | 粘贴完整的 SKILL.md 文件内容 |
| 模型（留空使用默认） | 文本 | 选填 |
输出
你将获得
以下产物及对应任务状态。
primarytext
使用说明
- 每行粘贴一个 SKILL.md 的完整内容。
- 系统将分析 skill 结构并生成对应的 TemplateSpec v2 JSON。
- 如果 skill 依赖工具调用，会返回错误信息。
示例
- SKILL.md 内容
- --- name: code-review description: Review code changes and provide structured feedback --- # Code Review Skill Review the provided code changes and give structured feedback. ## Instructions 1. Read the code changes carefully 2. Analyze each file for: - Code quality and readability - Potential bugs or logic errors - Performance issues - Security vulnerabilities - Missing error handling 3. Provide structured feedback with severity levels 4. Suggest specific improvements with code examples ## Output Format For each issue found: - **File**: filename - **Line**: line number(s) - **Severity**: Critical / Warning / Info - **Issue**: description - **Suggestion**: how to fix ## Rules - Be specific and actionable - Prioritize critical issues first - Include positive observations too
- SKILL.md 内容
- --- name: gif-sticker-maker description: Create animated GIF stickers from text descriptions --- # GIF Sticker Maker Create fun animated GIF stickers from text prompts. ## Input - Text description of the sticker (e.g., "a happy cat waving") - Style: cute / cool / funny / dramatic - Size: small (128x128) / medium (256x256) / large (512x512) ## Output - Animated GIF sticker file ## Process 1. Parse the text description 2. Generate key frames (4-8 frames) 3. Apply style effects 4. Assemble into animated GIF 5. Optimize file size ## Style Guidelines - Cute: pastel colors, rounded shapes, sparkles - Cool: dark theme, neon accents, sharp edges - Funny: exaggerated expressions, bright colors - Dramatic: high contrast, motion blur, epic lighting
- SKILL.md 内容
- --- name: vision-analysis description: Analyze uploaded images with AI vision capabilities --- # Vision Analysis Analyze uploaded images and provide detailed descriptions. ## Input - Image file (jpg, png, webp, gif) - Optional: analysis focus area ## Analysis Types ### Object Detection Identify and list all objects in the image with bounding boxes. ### Scene Understanding Describe the overall scene, context, and relationships between elements. ### Text Extraction (OCR) Extract all visible text from the image. ### Color Analysis Identify dominant colors and color palette. ## Output Format ```markdown ## Image Analysis Report ### Overview [Scene description] ### Objects Detected - Object 1: [description] (confidence: X%) - Object 2: [description] (confidence: X%) ### Text Found [Extracted text if any] ### Color Palette - Primary: #XXXXXX - Secondary: #XXXXXX - Accent: #XXXXXX ### Notes [Any additional observations] ```
可信信息与更新记录
当前版本v10
下载量2
调用量1
| 日期 | 版本 |
|---|---|
| 2026年9月4日 | v10 |
| 2026年8月26日 | v1 |
| 2026年8月26日 | v1 |