Video AI for Openclaw

An AI-powered video editing skill that transforms raw footage into professional 1080p MP4 videos using cloud GPU rendering.

tk8544-b
v1.0.0
Apr 11, 2026
0
170
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install video-ai

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install video-ai 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 AI?

The Video AI skill is a specialized tool within the collection of Openclaw Skills designed to simplify the post-production process for content creators and marketers. It allows users to send raw video footage in formats like MP4, MOV, or WebM and receive a fully edited, high-definition result without needing local editing software or powerful hardware.

By offloading processing to remote GPU nodes, this skill performs complex tasks such as silence removal, captioning, and multi-track compositing. It bridges the gap between raw recordings and professional-grade exports, making it an essential component for anyone looking to scale their video content production via Openclaw Skills.

Video AI Use Cases

  • Automatically editing and enhancing raw video recordings for social media.
  • Cutting silences from long screen recordings to create concise tutorials.
  • Adding AI-generated captions and text overlays to increase viewer engagement.
  • Batch processing multiple clips in a single session for consistent branding and style.

How Video AI Works

  1. The skill initializes by connecting to the backend API, either using a provided NEMO_TOKEN or acquiring an anonymous starter token.
  2. Users upload raw video files (up to 500MB) which are stored in a session-specific cloud environment.
  3. Natural language prompts are sent via an SSE (Server-Sent Events) stream, where the AI interprets requests like 'add BGM' or 'cut silences'.
  4. The cloud render pipeline processes these instructions on GPU nodes to composite layers and apply H.264 compression.
  5. The skill polls the render status every 30 seconds and provides a final download URL for the 1080p MP4 file once complete.

Video AI Setup

To get started with this skill, ensure you have your environment variables configured. If you have a specific API token, set it in your environment:

export NEMO_TOKEN="your_nemo_api_token"

For users without a token, the skill will automatically generate a UUID and acquire a free starter token with 100 credits. Ensure the configuration directory exists on your system:

mkdir -p ~/.config/nemovideo/

Video AI Data Schema & Taxonomy

The skill organizes video data using a structured session state that tracks the timeline and media metadata:

Key Description
session_id Manages the continuity of the video project across multiple requests.
draft The central JSON object representing the timeline, including tracks and segments.
tt Track type indicator: 0 for video, 1 for audio, and 7 for text/captions.
d Duration of segments measured in milliseconds (ms).
video_infos Metadata regarding the uploaded raw assets, such as resolution and bit rate.

Video AI Advanced Features

  • Multi-track editing support including background music (BGM) and dynamic text overlays.
  • Iterative timeline refinement where the session state persists between multiple prompts for non-destructive editing.
  • Remote GPU rendering for 1080p MP4 exports in under two minutes, optimized for fast turnaround.
  • Automatic credit balance management and anonymous session handling for instant access to Openclaw Skills capabilities.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Github Stars: 0
forks: 0

Featured*