An intelligent video processing tool that extracts keyframes and generates vision-based analysis reports with representative screenshots.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install vidu-video-analyzer
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 vidu-video-analyzer 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 Analyzer?
The Video Analyzer is a specialized utility designed for the Openclaw Skills framework that enables AI agents to "see" and interpret video content. By leveraging ffmpeg and advanced vision models, it bridges the gap between raw video files and actionable text-based insights.
This skill is engineered for performance and cost-effectiveness, utilizing I-frame detection to identify the most significant moments in a video. It reduces redundant data processing, ensuring that only the most relevant visual information is passed to the AI for analysis, resulting in high-quality summaries and representative imagery.
Video Analyzer Use Cases
- Analyzing video content when users share files via messaging platforms like Feishu.
- Generating automated summaries of video themes without manual watching.
- Identifying and extracting the top 3 most representative screenshots for documentation.
- Processing common video formats like MP4, MOV, and AVI for rapid content auditing.
How Video Analyzer Works
- Video Ingestion: The system detects a video file input and saves it to a local inbound directory.
- Metadata Extraction: It uses ffmpeg to parse duration, resolution, and codec information to prepare for processing.
- Intelligent Keyframe Extraction: A specialized script identifies I-frames to capture meaningful scene changes while maintaining token efficiency.
- Vision Analysis: The extracted frames are sent to a vision-capable LLM to identify themes and select the most important frames.
- Report Generation: A structured Markdown report is compiled, linking specific frames to thematic justifications.
- Result Dispatch: The final analysis and selected screenshots are sent back to the user interface.
Video Analyzer Setup
To install and configure the Video Analyzer within your Openclaw Skills environment, ensure you have ffmpeg installed and follow these steps:
# Ensure ffmpeg is available on your system
sudo apt update && sudo apt install ffmpeg
# Navigate to your local Openclaw directory
cd ~/.openclaw/workspace/skills/
# The skill utilizes the following script for extraction
bash ~/.openclaw/workspace/skills/video-analyzer/scripts/extract_keyframes.sh <video_path>
Video Analyzer Data Schema & Taxonomy
The skill manages video data and generated assets using a structured file hierarchy:
| Directory Path | Content Description |
|---|---|
~/.openclaw/media/inbound/ |
Raw video files received from users. |
~/.openclaw/media/keyframes/ |
Extracted JPEG images (640px width) for AI processing. |
~/.openclaw/workspace/skills/video-analyzer/scripts/ |
Shell scripts for I-frame detection and extraction. |
~/.openclaw/workspace/skills/video-analyzer/references/ |
Documentation and ffmpeg command references. |
Video Analyzer Advanced Features
- I-Frame Detection: Optimizes token consumption by ~7% compared to uniform sampling by only selecting high-value frames.
- Vision-Model Integration: Deep thematic analysis capable of distinguishing complex visual narratives.
- Multi-Format Support: Broad compatibility with MP4, MOV, and AVI formats.
- Automated Reporting: Generates formatted Markdown tables comparing frames and their thematic significance.
SKILL.md
Loading
Related Openclaw Skills
API Test Automation
A professional-grade API testing suite supporting REST and GraphQL with integrated performance, contract testing, and mock services.
Agent Spectrum
A standardized six-axis scoring framework designed for AI agents to evaluate their own capabilities or those of other agents with visual profiling.
SemVer Toolkit
A zero-dependency Python utility for parsing, validating, and manipulating Semantic Versioning 2.0.0 strings.
HPC ElmerFEM
A specialized agent skill for building, debugging, and automating complex ElmerFEM multiphysics simulation workflows and HPC execution.
Actuator Sizing and Selection Calculator
An industrial-grade CLI tool designed for precise actuator selection and sizing calculations with integrated data management.
CNC Machining Program Manager
A specialized CLI utility designed to organize, search, and manage industrial CNC machining programs and metadata.
Featured*
