Site Builder
apps site-builder tool revenue
Google Maps business listing to deployed React website in 60 seconds. Scrape, generate AI content, build, deploy, edit inline.
Architecture
- Frontend: Vue 3 + TypeScript + Pinia + Tailwind (port 5177)
- Backend: FastAPI + Python 3.13 + asyncio (port 9405)
- Generated sites: React 18 + Tailwind CSS (Vite build)
- AI Content: Claude Sonnet for copy, Gemini 2.5 Flash for images
- Deploy: Cloudflare Pages (primary), Vercel (fallback)
- Real-time: WebSocket for pipeline progress
Pipeline
| Step | Module | Output |
|---|---|---|
| 1. Parse Google Maps URL | maps_url_parser.py | ParsedMapsUrl |
| 2. Scrape business profile | maps_scraper.py (Playwright) | BusinessData |
| 3. Scrape business website | website_scraper.py (optional) | WebsiteData |
| 4. Generate AI content | site_generator.py (Claude) | SiteContent |
| 5. Generate AI images | image_generator.py (Gemini) | ImageGenerationResult |
| 6. Build React site | react_builder.py (npm build) | dist/ |
| 7. Deploy | cloudflare_deployer.py / vercel_deployer.py | Live URL |
Location
/home/spotcircuit/agentic/agent-experts/apps/site_builder/
Standalone repo: spotcircuit/site-builder (auto-synced via GitHub Actions)
Revenue Connection
This is a demonstration tool for the AI Integration Consulting and Knowledge Base Builder revenue streams. Shows end-to-end AI pipeline from scrape to deploy.
Related
- cloudflare-pages-deploy — deployment pattern with project limit management
- websocket-progress-pattern — real-time pipeline progress via WebSocket
- inline-editor-pattern — post-generation editing with live preview
- ai-content-pipeline — Claude + Gemini hybrid content generation
- act-learn-reuse-testing — self-improving test pattern
Source: raw/site-builder-readme.md, raw/site-builder-architecture.md | Ingested: 2026-04-08