Agent-Skills.md

Agent Skills: AI Video Generation

AI video generation using Google Veo or OpenAI Sora. Use when user wants to generate, create, or make videos from text prompts.

UncategorizedID: thrownlemon/claude-code-plugins/video-generation

Repository

thrownlemon/claude-code-plugins
ThrownLemonLicense: MIT
1

Install this agent skill to your local

pnpm dlx add-skill https://github.com/ThrownLemon/claude-code-plugins/tree/HEAD/plugins/video-gen/skills/video-generation

Skill Files

Browse the full folder contents for video-generation.

Download Skill

Loading file tree…

plugins/video-gen/skills/video-generation/SKILL.md

Skill Metadata

Name
video-generation
Description
AI video generation using Google Veo or OpenAI Sora. Use when user wants to generate, create, or make videos from text prompts.

AI Video Generation

When the user wants to generate videos from text descriptions, delegate to the video-generator subagent.

When to Use

Use this skill when the user:

  • Asks to "generate a video" or "create a video"
  • Wants to "animate" something
  • Mentions "video of [something]"
  • Wants AI video generation from a prompt
  • Asks for text-to-video conversion

What the Subagent Handles

The video-generator subagent will:

  1. Check API Keys

    • Verify GOOGLE_API_KEY (for Veo)
    • Verify OPENAI_API_KEY (for Sora)
    • Guide setup if needed

  2. Select Provider

    • Smart routing based on request
    • Consider duration and resolution needs
    • Use available API keys

  3. Generate Video

    • Submit to selected API
    • Poll for completion
    • Download and save result

  4. Report Progress

    • Show generation status
    • Report completion time
    • Provide output path

Supported Providers

Google Veo (via Gemini API): | Model | Full ID | Status | Audio | |-------|---------|--------|-------| | Veo 3.1 | veo-3.1-generate-preview | Preview | Yes | | Veo 3.1 Fast | veo-3.1-fast-generate-preview | Preview | Yes | | Veo 3.0 | veo-3.0-generate-001 | Stable | Yes | | Veo 3.0 Fast | veo-3.0-fast-generate-001 | Stable | Yes | | Veo 2.0 | veo-2.0-generate-001 | Stable | No |

  • Duration: 4, 6, 8 seconds
  • Resolution: 720p, 1080p, 4K
  • Aspect: 16:9, 9:16

OpenAI Sora (⚠️ being discontinued — see deprecation below): | Model | ID | Use Case | |-------|-----|----------| | Sora 2 | sora-2 | Fast iteration, exploration | | Sora 2 Pro | sora-2-pro | Production quality |

Snapshot variants available: sora-2-2025-10-06, sora-2-2025-12-08, sora-2-pro-2025-10-06

⚠️ Sora Deprecation Notice: OpenAI is shutting down the Sora product. The consumer app went dark on April 26, 2026, and the /v1/videos API will be discontinued on September 24, 2026. There is no official OpenAI replacement. After that date, this plugin's OpenAI/Sora path will stop working — switch to the Google Veo path or a third-party provider (Runway, Pika, etc.).

  • Duration: 4, 8, 12 seconds
  • Aspect: 16:9, 9:16, 1:1
  • Audio: Included in output

Default Options

| Option | Veo Default | Sora Default | |--------|-------------|--------------| | Duration | 6 seconds | 8 seconds | | Aspect | 16:9 | 16:9 | | Resolution | 1080p | 1920x1080 | | Model | veo-3.1-generate-preview | sora-2 |

Provider Selection: Auto-selected based on:

  1. Available API keys
  2. Requested features (e.g., 12s duration requires Sora)
  3. User preference if specified

Example Triggers

  • "Generate a video of a sunset over mountains"
  • "Create a video showing a robot dancing"
  • "Make a video of autumn leaves falling"
  • "I need a video of abstract flowing colors"
  • "Animate this scene: a cat playing with yarn"
  • "Can you create a short video of fireworks?"