Analyzer- Primary vision analysis interfaceVideoProcessor- Video analysis capabilities
Rect- Bounding box rectangleSize- Image dimensionsUIElement- Detected UI elementTextRegion- Detected text regionVisualIssue- Detected visual problemAction- Navigation/interaction actionScreenIdentity- Unique screen identifierScreenAnalysis- Full screen analysis resultScreenDiff- Screen comparison resultKeyFrame- Video key frame
NewStubAnalyzer()- Create stub analyzer (no OpenCV)NewStubAnalyzerWithProvider(provider)- Create stub with LLM vision
NavigationGraph- Directed graph of screens and transitions
ScreenNode- Node in the graphTransition- Directed edgeGraphSnapshot- Serializable graph state
NewNavigationGraph()- Create empty graphExportDOT(g)- Export to Graphviz DOTExportJSON(g)- Export to JSONExportMermaid(g)- Export to Mermaid
VisionProvider- LLM vision API adapter
ProviderConfig- Provider configurationFallbackProvider- Multi-provider with fallback
NewOpenAIProvider(config)- GPT-4o adapterNewAnthropicProvider(config)- Claude adapterNewGeminiProvider(config)- Gemini adapterNewQwenProvider(config)- Qwen-VL adapterNewFallbackProvider(providers...)- Fallback chain
Config- VisionEngine configuration
DefaultConfig()- Sensible defaultsLoadFromEnv()- Load from environment variables