The Video Create skill automates the production of 1080p MP4 videos from raw images and clips using cloud-based GPU rendering and AI workflows.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install video-create
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install video-create using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Video Create?
The Video Create skill is a professional-grade tool designed for AI agents to bridge the gap between static assets and high-quality video content. By leveraging the power of Openclaw Skills, it allows users to upload images or short clips and transform them into finished 1080p MP4 files without the need for local video editing software. The skill manages the entire pipeline—from asset upload and timeline management to background music integration and final rendering—on high-performance cloud GPUs.
Built for developers and content creators, this skill interprets natural language commands to perform complex editing tasks like transitions, text overlays, and audio syncing. Whether you are building a marketing bot or a creative assistant, this skill provides the necessary infrastructure to generate cinematic results in under two minutes, handling everything from session management to status polling automatically.
Video Create Use Cases
- Creating 30-second promotional videos from a set of product photos and a company logo.
- Converting travel snapshots into a high-definition 1080p MP4 highlight reel with background music.
- Automating the generation of social media content by combining short clips and descriptive text overlays.
- Iterative video editing where users can refine a draft through natural language feedback within the AI agent chat.
How Video Create Works
- The skill initializes a connection to the backend API, authenticating via a NEMO_TOKEN or generating a temporary anonymous session identifier.
- Assets such as PNG, JPG, or MP4 files are uploaded to the cloud render pipeline, where they are staged for processing.
- The AI agent sends editing instructions through a Server-Sent Events (SSE) stream, allowing for real-time updates to the video timeline.
- The cloud GPU node composites the various layers, applying transitions and platform-specific H.264 compression.
- The system monitors the rendering progress by polling the cloud proxy until the status is marked as completed.
- The final 1080p MP4 file is made available via a secure download URL for the user.
Video Create Setup
To integrate this skill into your Openclaw Skills workflow, you need to configure the required environment variables. The skill can operate with a free starter token if a primary token is not provided.
# Set your Nemo Video API token for full access
export NEMO_TOKEN="your_token_here"
# The skill automatically manages configuration and session paths at:
# ~/.config/nemovideo/
Ensure that your requests include the necessary attribution headers (X-Skill-Source, X-Skill-Version) to avoid 402 errors during the export phase.
Video Create Data Schema & Taxonomy
The skill organizes video data using a structured draft schema within the session state:
| Key | Description |
|---|---|
session_id |
Unique identifier for the active video project. |
draft |
The core JSON object containing the timeline and track segments. |
t (Tracks) |
Array of layers including video (0), audio (1), and text (7). |
d (Duration) |
Integer value representing the length of segments in milliseconds. |
generated_media |
Metadata tracking all AI-generated assets within the session. |
credits |
Information regarding available render balance for the user account. |
Video Create Advanced Features
- Server-Sent Events (SSE) integration for real-time AI feedback and iterative timeline modification.
- High-performance cloud GPU rendering supporting 1080p output at 1080x1920 or 1920x1080 resolutions.
- Automated asset synchronization that maps natural language prompts (e.g., "add music", "make it 30 seconds") to internal API commands.
- Cross-platform attribution support for Openclaw Skills deployments in environments like Clawhub or Cursor.
- Comprehensive error handling and session recovery for long-running rendering tasks.
SKILL.md
Loading
Related Openclaw Skills
cms-docdb
A professional-grade AI skill for managing enterprise knowledge bases with robust version control and file operations.
Setup OpenViking Remote
An automated agent skill that installs and configures the OpenViking memory plugin for OpenClaw using natural language conversation.
Sun-Panel
A powerful AI-driven management skill for Sun-Panel navigation dashboards on NAS and home servers.
Accenture Business Intelligence Skill
A technical knowledge repository for analyzing Accenture's global consulting operations, service models, and digital transformation strategies.
Claimback Radar
An AI-powered scanner that analyzes emails and bills to identify subscription traps, hidden refunds, and savings opportunities.
SkVM JIT Optimize Trigger
A specialized tool to trigger just-in-time optimization for AI agent skills based on real-world task performance and failure logs.
Featured*
