Video Analysis

AI-powered video content understanding capability that enables intelligent analysis of video files.

Prerequisites

•Python 3.7+
•ARK_API_KEY environment variable (Volces ARK API required)
•Required packages: openai (Volces ARK API is compatible with OpenAI SDK)

Note: This skill uses Volces (火山引擎) ARK API for video analysis. The API is OpenAI-compatible, so we use the OpenAI SDK with a custom base URL.

Quick Start

bash

# Set up API key
export ARK_API_KEY="your_api_key_here"

# Run analysis
python -c "
from scripts.video_analyzer import VideoAnalyzer
result = VideoAnalyzer().analyze_video('demo.mp4', 'general')
print(result['analysis_result']['summary'])
"

Common Use Cases

Scenario	Recommended Type	Example
Quick video overview	`general`	"Analyze this video and give me a summary"
Product demo analysis	`product`	"Extract features from this app demo"
Test case extraction	`key_nodes`	"Find key moments for QA testing"
UI/UX review	`custom`	"Analyze accessibility issues in this video"
Documentation	`key_nodes` + `general`	"Create documentation from tutorial video"
Competitive analysis	`product`	"Understand competitor's features"

Analysis Types

Type	Use Case	Command
`general`	Quick content overview and summary	`analyze_video(path, "general")`
`product`	App/feature understanding from product perspective	`analyze_video(path, "product")`
`key_nodes`	Timestamped moments for testing/documentation	`analyze_video(path, "key_nodes")`
`custom`	Specialized analysis with your prompt	`analyze_with_prompt(path, custom_prompt)`

See Analysis Types for detailed usage of each type.

Complete Workflow Example

python

from scripts.video_analyzer import VideoAnalyzer

# Initialize analyzer
analyzer = VideoAnalyzer()

# Example 1: Get video summary
result = analyzer.analyze_video("demo.mp4", "general")
print(result['analysis_result']['summary'])

# Example 2: Extract product features
result = analyzer.analyze_video("app_demo.mp4", "product")
print(f"Product: {result['analysis_result']['product_name']}")
print(f"Features: {result['analysis_result']['key_features']}")

# Example 3: Get key moments for testing
result = analyzer.analyze_video("tutorial.mp4", "key_nodes")
for node in result['analysis_result']['key_nodes']:
    print(f"[{node['timestamp']}] {node['description']}")

See API Reference for complete documentation.

Output Format

All results return structured JSON with analysis_result and metadata.

See Output Guide for quality standards and examples.

Error Handling

Handle common errors: FileNotFoundError, ValueError, RuntimeError.

See Error Handling for detailed troubleshooting.

Configuration

Set environment variable:

bash

# Option 1: Temporary (current session only)
export ARK_API_KEY="your_api_key_here"

# Option 2: Permanent (add to ~/.zshrc)
echo 'export ARK_API_KEY="your_api_key_here"' >> ~/.zshrc
source ~/.zshrc

# Verify it's set
echo $ARK_API_KEY

Optional: Override in VideoAnalyzer constructor with api_key, base_url, or model parameters.