An advanced video analysis skill providing high-level technical quality assessments and granular shot-by-shot narrative breakdowns.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install video-breakdown
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-breakdown 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 Breakdown?
The Video Breakdown skill is a sophisticated tool designed for content creators, editors, and digital marketers who require objective, data-driven evaluations of video content. By acting as a specialized client for backend services powered by fal.ai, this skill enables deep deconstruction of media files into structured data. Whether you need to assess the technical fidelity of a clip or understand the cinematic structure of a scene, this integration within the Openclaw Skills ecosystem provides professional-grade results.
This tool bridges the gap between raw video files and actionable insights, allowing users to automate the review process. It offers two primary modes: a quantitative Quality Critique that scores technical attributes and a qualitative Shot Breakdown that describes visual transitions and camera movements in detail.
Video Breakdown Use Cases
- Benchmarking user-generated content against professional quality standards before distribution.
- Deconstructing competitor video content to analyze pacing, camera angles, and narrative flow.
- Generating automated scripts or metadata for highlight reels based on AI-generated shot descriptions.
- Performing technical audits on video libraries to identify assets with poor lighting or stability issues.
How Video Breakdown Works
- The agent initiates the process by calling the /analyze endpoint with a target video URL and a specific analysis type.
- The backend service receives the request, queues the job via high-performance AI models, and returns a unique task_id.
- The agent enters a polling phase, periodically checking the /status/{task_id} endpoint to monitor progress.
- Once the AI model completes the analysis, the service transitions the status to completed.
- The agent retrieves the final structured JSON result and presents the breakdown or critique to the user.
Video Breakdown Setup
To integrate this capability into your environment, ensure your agent is configured to communicate with the provided backend API. Openclaw Skills work best when the backend service is properly authenticated and reachable.
- Configure your environment variables to point to the video-breakdown service endpoint.
- Verify the available analysis types by querying the metadata endpoint:
curl -X GET https://api.your-backend.com/analysis_types
- Start a test analysis by submitting a video URL:
curl -X POST https://api.your-backend.com/analyze \
-H "Content-Type: application/json" \
-d '{"video_url": "https://example.com/video.mp4", "analysis_type": "quality_critique"}'
Video Breakdown Data Schema & Taxonomy
The skill returns structured JSON objects that vary based on the analysis_type selected. This consistency allows Openclaw Skills to pass data seamlessly to other automation tools.
| Feature | Data Format | Description |
|---|---|---|
| Technical Scores | Object | Includes scores (1-10) and comments for resolution, lighting, audio, and stability. |
| Scene Metadata | Array | A list of shots containing start/end timestamps and visual descriptions. |
| Camera Work | String | Detailed notes on camera angles, zooms, and movement within each shot. |
| Status Tracking | String | Current state of the job: queued, in_progress, or completed. |
Video Breakdown Advanced Features
- Asynchronous task management allows for the analysis of long-form video content without connection timeouts.
- Support for shot-by-shot 'La Pian' (拉片) analysis, providing cinematic-level detail for education and critique.
- Quantitative scoring system that enables programmatic filtering of high-quality vs. low-quality assets.
- Multi-modal output combining technical metrics with natural language visual descriptions.
SKILL.md
Loading
Related Openclaw Skills
Transition Design
An AI-powered design tool that analyzes clip boundaries and narrative intent to generate precise, frame-accurate video transition instructions.
SimilarWeb Analytics
A comprehensive domain analysis tool providing deep insights into traffic, engagement, and global rankings using SimilarWeb data.
SFX Generator
A powerful AI skill for generating custom sound effects from text prompts using the ElevenLabs API.
Seedance Prompt Designer
An intelligent prompt engineering tool that transforms multimodal assets and creative intent into structured video generation prompts for the Seedance 2.0 model.
Video Upscaler
A multi-model AI video enhancement tool that transforms low-resolution or noisy footage into cinematic 4K quality.
Skill Publisher
A dedicated automation tool for the multi-channel distribution and lifecycle management of AI agent skills.
Featured*
