@unityclaw/skills 1.0.0

package/unityclaw-video-generation-veo/SKILL.md

@@ -0,0 +1,239 @@
+---
+name: unityclaw-video-generation-veo
+description: Generate high-quality videos using Google Veo AI model
+version: 1.0.0
+metadata:
+  openclaw:
+    requires:
+      env:
+        - UNITYCLAW_API_KEY
+      bins:
+        - node
+        - npm
+    primaryEnv: UNITYCLAW_API_KEY
+    emoji: "🎥"
+    homepage: https://unityclaw.com
+    install:
+      - kind: node
+        package: "@unityclaw/sdk"
+        bins: []
+---
+
+# UnityClaw Video Generation - Veo
+
+Generate high-quality videos using Google's Veo AI model with advanced control over aspect ratio, resolution, and duration.
+
+## Installation
+
+```bash
+npm install @unityclaw/sdk
+```
+
+## Configuration
+
+Set your API key as environment variable:
+
+```bash
+export UNITYCLAW_API_KEY=your-api-key
+```
+
+## Quick Start
+
+```typescript
+import { UnityClawClient } from '@unityclaw/sdk';
+
+const client = new UnityClawClient();
+
+const result = await client.video.veo({
+  prompt: 'A beautiful sunset over the ocean with waves crashing',
+  aspect_ratio: { value: '16:9', label: '16:9' },
+  duration: { value: '8', label: '8s' }
+});
+
+if (result.code === 0) {
+  console.log('Generated video:', result.data);
+}
+```
+
+## API Reference
+
+### veo()
+
+Generate video using Google Veo model.
+
+```typescript
+await client.video.veo({
+  prompt?: string | TextFieldItem[];
+  attachment?: AttachmentFieldItem[];
+  first_frame?: AttachmentFieldItem[];
+  last_frame?: AttachmentFieldItem[];
+  aspect_ratio?: LabelFieldItem | string;
+  resolution?: LabelFieldItem | string;
+  duration?: LabelFieldItem | string;
+}): Promise<APIResponse<AttachmentResult[]>>
+```
+
+## Parameters
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| `prompt` | `string \| TextFieldItem[]` | No* | Text description (*required if no attachment) |
+| `attachment` | `AttachmentFieldItem[]` | No* | Reference image (*required if no prompt) |
+| `first_frame` | `AttachmentFieldItem[]` | No | Starting frame for video |
+| `last_frame` | `AttachmentFieldItem[]` | No | Ending frame for video |
+| `aspect_ratio` | `LabelFieldItem \| string` | No | Video aspect ratio |
+| `resolution` | `LabelFieldItem \| string` | No | Video resolution |
+| `duration` | `LabelFieldItem \| string` | No | Video duration in seconds |
+
+## Examples
+
+### Text-to-Video
+
+```typescript
+const client = new UnityClawClient();
+
+const result = await client.video.veo({
+  prompt: 'A majestic eagle soaring through misty mountain peaks at sunrise',
+  aspect_ratio: { value: '16:9', label: '16:9' },
+  resolution: { value: '1080p', label: '1080p' },
+  duration: { value: '8', label: '8s' }
+});
+```
+
+### Image-to-Video
+
+```typescript
+const result = await client.video.veo({
+  attachment: [
+    { tmp_url: 'https://example.com/landscape.jpg', name: 'landscape.jpg' }
+  ],
+  aspect_ratio: { value: '16:9', label: '16:9' },
+  duration: { value: '5', label: '5s' }
+});
+```
+
+### First/Last Frame Control
+
+```typescript
+// Generate video with specific start and end frames
+const result = await client.video.veo({
+  first_frame: [
+    { tmp_url: 'https://example.com/start.jpg', name: 'start.jpg' }
+  ],
+  last_frame: [
+    { tmp_url: 'https://example.com/end.jpg', name: 'end.jpg' }
+  ],
+  prompt: 'Smooth transition between the two scenes',
+  duration: { value: '8', label: '8s' }
+});
+```
+
+### High Resolution Video
+
+```typescript
+const result = await client.video.veo({
+  prompt: 'A cinematic shot of a luxury car driving through a city at night',
+  aspect_ratio: { value: '16:9', label: '16:9' },
+  resolution: { value: '4K', label: '4K' },
+  duration: { value: '10', label: '10s' }
+});
+```
+
+### Portrait Video
+
+```typescript
+const result = await client.video.veo({
+  prompt: 'A model walking down a fashion runway with dramatic lighting',
+  aspect_ratio: { value: '9:16', label: '9:16' },
+  duration: { value: '5', label: '5s' }
+});
+```
+
+### Simple String Parameters
+
+```typescript
+// You can also use simple strings instead of LabelFieldItem
+const result = await client.video.veo({
+  prompt: 'A peaceful garden scene',
+  aspect_ratio: '16:9',
+  resolution: '1080p',
+  duration: '8'
+});
+```
+
+## Aspect Ratio Options
+
+| Value | Use Case |
+|-------|----------|
+| `16:9` | Standard landscape (YouTube, presentations) |
+| `9:16` | Portrait (TikTok, Instagram Reels) |
+| `1:1` | Square (Instagram feed) |
+
+## Duration Options
+
+| Value | Description |
+|-------|-------------|
+| `5` | 5 seconds |
+| `8` | 8 seconds (default) |
+| `10` | 10 seconds |
+
+## Resolution Options
+
+| Value | Description |
+|-------|-------------|
+| `720p` | HD quality |
+| `1080p` | Full HD |
+| `4K` | Ultra HD |
+
+## Response Format
+
+```typescript
+interface AttachmentResult {
+  name: string;
+  contentType: 'attachment/url';
+  content: string;  // URL to generated video
+}
+```
+
+## Error Handling
+
+```typescript
+const result = await client.video.veo({
+  prompt: 'A test video',
+  aspect_ratio: '16:9'
+});
+
+if (result.code !== 0) {
+  console.error('Video generation failed:', result.msg);
+} else {
+  console.log('Success:', result.data);
+}
+```
+
+## Task Folders
+
+Each execution creates a task folder:
+
+```typescript
+const result = await client.video.veo({
+  prompt: 'test',
+  aspect_ratio: '16:9'
+});
+
+console.log('Task ID:', result.taskId);
+console.log('Task Folder:', result.taskFolder);
+console.log('Downloaded Videos:', result.attachments);
+```
+
+## Best Practices
+
+1. **First/Last Frame**: Use for precise control over video start/end
+2. **Resolution**: Match to your output requirements
+3. **Duration**: Shorter videos generate faster
+4. **Prompts**: Be specific about camera movement and style
+
+## Related Skills
+
+- [unityclaw-video-generation-sora](../unityclaw-video-generation-sora/SKILL.md) - OpenAI Sora video generation
+- [unityclaw-video-generation-kling](../unityclaw-video-generation-kling/SKILL.md) - Kling video generation
+- [unityclaw-image-generation](../unityclaw-image-generation/SKILL.md) - Generate reference images