veo-video-generator

What This Skill Does

The veo-video-generator skill integrates Google's advanced Veo 3.1 model directly into your OpenClaw environment, allowing you to produce high-fidelity 1080p video clips with integrated, synced audio. Designed for creators and developers, this skill abstracts the complex API interactions with the Google GenAI infrastructure, letting you focus on creative output. It handles the rendering process in a secure, isolated manner, producing professional-grade 9:16 aspect ratio clips that are ideal for social media or cinematic prototypes.

Installation

To integrate this skill, ensure you have the OpenClaw agent initialized. Open your terminal or the OpenClaw console and run the command: clawhub install openclaw/skills/skills/kghamilton89/veo-video-generator. Once installed, you must perform a one-time dependency initialization by executing npm install within the skill directory to fetch the necessary client libraries. Ensure that the GEMINI_API_KEY is correctly set in your environment variables, as the skill requires this credential to authenticate with the Google GenAI API.

Use Cases

This skill is perfect for rapid prototyping in film production, creating engaging content for social media platforms, or adding dynamic visual elements to technical presentations. It allows users to turn abstract concepts or written screenplays into tangible video assets without the need for high-end rendering software or complex video editing tools.

Example Prompts

1. "Generate a 1080p cinematic video of a futuristic cyberpunk city street during a light rain, with ambient neon reflections and synthesized electronic music."
2. "Create a short video clip showing a hyper-realistic close-up of a succulent plant unfolding its leaves in time-lapse, including nature sound effects."
3. "Render a video of a golden retriever playing fetch on a beach at sunset, ensuring the audio captures the sound of crashing waves and joyful barking."

Tips & Limitations

For the best results, use descriptive, vivid language in your prompts to help the model better understand the aesthetic, lighting, and mood requirements. Note that this skill defaults to a 9:16 aspect ratio, which is optimized for vertical video platforms. Always ensure your prompts are passed as discrete arguments to the execution function to prevent shell injection vulnerabilities. Remember that the skill requires an active internet connection to access the Google GenAI cloud endpoints, and video generation times may vary based on server load and complexity of the requested scene.