VibeFast
Customizing Features

AI Image Analysis

Add camera-driven analysis flows that stream photos to your backend for AI scoring.

This feature (registered as image-analysis in the CLI) wires a camera-first workflow, backend storage, and configurable scoring traits so you can analyze faces, dashboards, or any visual input you choose.

vf add image-analysis

Frontend flow

  • Route: apps/native/src/app/(root)/(protected)/image-analysis renders the capture and results screens.
  • Master config: apps/native/src/features/image-analysis/config/master-analysis-config.ts defines each flow (face analysis, dashboard warning) with its icon, color, and photo steps.
  • Screens: apps/native/src/features/image-analysis/app contains the capture screen, options screen, results screen, and helpers to load the configs.
  • Components: apps/native/src/features/image-analysis/components handle achievements, galleries, and results cards that reuse the theme tokens.

Backend flow

  • packages/backend/convex/imageAnalysis.ts stores uploads, queues AI analysis, and updates the imageAnalyses table
  • generateImageAnalysis in packages/backend/convex/lib/ai/imageAnalysisAdapter calls your AI provider
  • packages/backend/supabase/functions/analyze-image/ Edge Function calls your AI provider
  • packages/backend/src/services/image-analyses.ts handles results storage
  • Database table: image_analyses with RLS policies

Configuration

  • Update packages/backend/.env.local with AI_PROVIDER, AI_MODEL, OPENAI_API_KEY, GEMINI_API_KEY
  • Your backend must be running (pnpm dev:server) while you test

Customization

  • Add new analysis configs by extending MASTER_ANALYSIS_CONFIGS
  • Each backend config defines traits, prompts, and a schema for validation
  • The UI automatically binds to photo steps and available goals

Found an issue or bug in the docs?

Help me improve! If you spot any errors, typos, or have suggestions, please let me know.

Reach out on X/Twitter @zafarbuildzz

On this page

Frontend flowBackend flowConfigurationCustomization