An AI-driven toolkit for extracting text, scenes, and transcriptions from video advertisements using Google Gemini Vision.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install meta-video-ad-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 meta-video-ad-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 Ad Analyzer?
The Video Ad Analyzer is a sophisticated tool designed to break down video creative content into actionable data. By leveraging Google Gemini Vision AI, it performs frame extraction with scene change detection, optical character recognition (OCR), and audio transcription to provide a comprehensive view of any video ad. This skill within the Openclaw Skills ecosystem allows developers and marketers to programmatically understand video content, identifying key visuals and messaging without manual review.
By synthesizing visual and auditory data, the tool generates a complete timeline of a video's narrative. Whether you are performing competitive analysis or automating content tagging, this addition to your Openclaw Skills library provides the technical depth needed for high-scale video processing.
Video Ad Analyzer Use Cases
- Analyzing video creative content to identify high-performing visual hooks and scene transitions.
- Extracting text overlays and calls-to-action (CTAs) for competitive marketing research.
- Generating automated scene-by-scene descriptions for content accessibility or digital asset management.
- Transcribing video ad audio to analyze verbal messaging and keyword density.
- Batch generating thumbnails and metadata for large-scale video libraries.
How Video Ad Analyzer Works
- The system utilizes histogram-based change detection to sample frames only when significant scene changes occur.
- EasyOCR scans the sampled frames to detect and extract visible text overlays at specific intervals.
- The audio track is isolated and processed through Google Cloud Speech-to-Text for a complete verbal transcript.
- Gemini Vision AI analyzes the visual frames to provide semantic descriptions of the action and environment.
- The Openclaw Skills logic reconciles OCR and AI Vision data to provide a unified, proofread output of all extracted information.
Video Ad Analyzer Setup
1. Environment Variables
# Required for Gemini Vision and Speech-to-Text
export GOOGLE_APPLICATION_CREDENTIALS="/path/to/service-account.json"
2. Dependencies
pip install opencv-python pillow easyocr ffmpeg-python google-cloud-speech vertexai google-api-python-client
Note: This skill requires ffmpeg and ffprobe to be installed on your system to handle media processing.
Video Ad Analyzer Data Schema & Taxonomy
The skill returns an ExtractedVideoContent object with the following structure:
| Field | Description |
|---|---|
| video_path | Local file path to the processed video |
| duration | Total video duration in seconds |
| transcript | Full text generated from audio transcription |
| text_timeline | Timestamped list of detected OCR text and overlays |
| scene_timeline | Timestamped list of AI-generated descriptions for each scene |
| thumbnail_url | Path to the automatically generated video thumbnail |
| extraction_complete | Boolean status of the processing job |
Video Ad Analyzer Advanced Features
- Smart frame sampling using histogram-based scene change detection with configurable thresholds.
- Tiered OCR confidence thresholds to ensure high-quality text extraction from complex backgrounds.
- AI proofreading where Gemini models reconcile raw OCR errors into coherent sentences.
- Support for native video analysis for files under 20MB, allowing direct Gemini processing.
- Highly customizable AI behavior via prompt files within the Openclaw Skills directory structure.
SKILL.md
Loading
Related Openclaw Skills
Meta Ad Creatives
A data-driven performance tracking skill for analyzing Meta Ads creative hit rates and cross-account benchmarks.
YouTube Idea Extractor
A specialized tool for mining YouTube transcripts to identify and validate high-potential startup opportunities and business ideas.
Video Promo Generator
Automate the creation of high-impact promo video plans, scripts, and tool-specific configurations directly from your project's codebase.
Solo-Stream Decision Framework
Solo-Stream is an interactive AI wizard that guides founders through the 6-layer STREAM framework to make high-stakes decisions with clarity and data.
Video Ad Deconstructor
An AI-powered analysis tool that deconstructs video ad creatives into actionable marketing insights like hooks, social proof, and emotional triggers.
Memory Manager
A sophisticated three-tier memory architecture providing sliding windows, automated summarization, and vector-based long-term storage for AI agents.
Featured*
