A powerful AI-driven search tool that finds and extracts specific moments from video clips based on your natural language descriptions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install video-google
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-google 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 Google?
Video Google is a sophisticated skill designed for Openclaw Skills that leverages a cloud rendering pipeline and AI search to eliminate the need for manual video scrubbing. Users can upload video files and query them using natural language, such as asking to find a specific mention of a topic or a particular visual event. The skill processes these requests through a high-performance GPU-backed pipeline, returning timestamped, 1080p MP4 clips ready for immediate use.
Built specifically for high-velocity workflows, this tool is ideal for those who need to navigate long recordings—like lectures, meetings, or raw footage—to find specific content without touching a single slider. By integrating directly into your developer environment, it transforms how you interact with media assets, making video editing as simple as typing a search query.
Video Google Use Cases
- Finding specific quotes or mentions of keywords (e.g., quarterly revenue) in long meeting recordings.
- Identifying and extracting key segments from academic lectures for research or study aids.
- Creating quick social media clips from raw footage by searching for specific actions or topics.
- Batch processing multiple video uploads to generate various cut-downs based on different instructions.
How Video Google Works
- The user uploads a video file (up to 500MB) or provides a URL to the Openclaw Skills interface.
- An automated setup routine connects to the Nemo Video API, generating a session and securing an authentication token.
- The user submits a natural language search query describing the moment they want to find.
- The request is sent via a Server-Sent Events (SSE) stream to a cloud GPU node that analyzes the video content.
- The cloud pipeline composites the identified layers and renders a 1080p MP4 using H.264 compression.
- The skill polls the render status every 30 seconds until completion and provides a direct download URL to the user.
Video Google Setup
To get started with Video Google, ensure your environment is configured. On the first interaction, the skill will handle most setup steps automatically.
# Set your API token as an environment variable
export NEMO_TOKEN="your_token_here"
# The skill will automatically attempt to connect
# and create a session on your first search request.
Video Google Data Schema & Taxonomy
The skill manages video data and session states using a structured JSON schema for timelines and track management.
| Field | Description |
|---|---|
task_id |
Unique identifier for the current video processing task. |
session_id |
Persists the timeline state for iterative editing. |
draft |
JSON object containing tracks (t), track types (tt), and durations (d). |
video_infos |
Metadata regarding uploaded source files (dimensions, codec). |
generated_media |
References to the final rendered outputs and download links. |
Video Google Advanced Features
- Multi-track timeline management supporting video, audio, and text overlays.
- Iterative editing within a single session, allowing for refinement of search results without re-uploading.
- Support for batch uploads and sequential processing of multiple files.
- Automatic conversion of GUI-style instructions into API-driven edit commands.
SKILL.md
Loading
Related Openclaw Skills
In One Video — AI Video Merger
An AI-powered video editing skill that merges multiple clips into a single high-quality video using natural language commands.
Pika AI Video
A high-performance AI video generation tool that converts text prompts and images into cinematic 1080p MP4 clips using cloud-based GPU rendering.
Video Editor ZH
An AI-powered video editing tool designed for Chinese content creators to automate subtitling, transitions, and 1080p exports.
Video AI Editor
Edit and export professional, polished videos using simple AI commands without manual sliders.
One Video AI
A cloud-based AI video editing skill that transforms raw footage into polished, high-definition videos through natural language commands.
Doorstep Video — AI Video Editing and Export
A specialized AI skill for transforming raw doorbell and doorstep footage into polished, shareable 1080p MP4 video clips using natural language commands.
Featured*
