A high-performance AI agent skill for generating cinema-quality video and audio-synced content using the Kling 3.0 Omni model.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install kling-video-generator
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install kling-video-generator using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Kling 3.0 Omni Video Generator is a robust technical implementation designed for the Openclaw Skills ecosystem, enabling users to generate, edit, and manipulate high-fidelity video content. It bridges the gap between raw user intent and the complex API requirements of the Kling 3.0 Omni model, providing a streamlined workflow for text-to-video, image-to-video, and sophisticated video-to-video transformations.
This skill is built for developers who need precise control over cinematic AI generation. By integrating this into your Openclaw Skills library, you gain access to a structured decision-making engine that handles everything from prompt engineering using official formulas to managing intricate parameter constraints like mutual exclusion rules and JWT-based authentication.
To deploy this within your environment, ensure you have Python 3 and the necessary API credentials from the Kling platform.
# Set your Kling API credentials in your environment
export KLING_ACCESS_KEY="your_access_key"
export KLING_SECRET_KEY="your_secret_key"
Ensure that the kling_api.py script is accessible within your project structure to leverage the full capabilities of these Openclaw Skills.
The skill manages data through a structured payload system and task-tracking metadata. The primary data points include:
| Attribute | Description | Requirement |
|---|---|---|
model_name |
Fixed to kling-v3-omni for high-quality output |
Required |
multi_prompt |
Array of shot objects containing prompt, index, and duration | Multi-shot only |
refer_type |
Specifies either base for editing or feature for style reference |
Required for Video-to-Video |
sound |
Toggle for synchronized audio generation | Incompatible with video_list |
image_list |
Collection of reference images (max 4-7 depending on mode) | Optional |
<<<image_1>>>, <<<video_1>>>) for explicit multimodal asset referencing.Loading
A versatile image generation skill that intelligently selects between Midjourney, Flux, and specialty models to create high-quality visuals.

A comprehensive image generation toolkit for Openclaw Skills that provides a unified interface for 8 industry-leading AI models including Midjourney and Flux.

An automated utility for upscaling, sharpening, and refining screenshots and digital images for professional use.

A specialized tool for finding and implementing battle-tested open-source GitHub projects to solve technical problems without reinventing the wheel.

A comprehensive toolset for analyzing stocks, company profiles, and financial market data through real-time technical and fundamental insights.

An automated execution engine that transforms structured composition plans into high-fidelity AI-generated audio assets.








































