Build a Menu Engineering Tool with AI

Build a static HTML menu engineering analysis tool for restaurant owners. Dark theme with background #0a0a0f, surfaces #12121a, borders #1a1a2a, accent color #dc2626 (red). All logic runs in-browser with inline JavaScript. Core feature: a data entry table where users input menu items with: item name, category, menu price, food cost ($), and units sold per week. The tool calculates contribution margin (price minus food cost), classifies each item using the Kasavana & Smith matrix into one of four quadrants — Star (high margin + high popularity), Plowhorse (low margin + high popularity), Puzzle (high margin + low popularity), Dog (low margin + low popularity) — and displays results as an interactive 2x2 SVG scatter plot. Below the chart, show a ranked table of all items color-coded by quadrant with action recommendations: Stars (keep, feature prominently), Plowhorses (reprice or reduce portion), Puzzles (reposition on menu, improve description), Dogs (remove or replace). Include a "Menu Mix %" column showing each item's share of total sales volume. Add a category breakdown view: toggle between Appetizers, Mains, Desserts, Drinks. Show a revenue summary bar: total weekly revenue, average contribution margin, and theoretical max if all Dogs were replaced with Stars. Seed 20 sample menu items across 4 categories with realistic pricing and cost data. Export results as a downloadable CSV. Read the guide at https://accessagent.ai/api/guide and follow it to deploy the site to AccessAgent.ai. My email is [your@email.com].

Update the menu engineering tool's sample data with new weekly sales figures reflecting summer seasonality (higher salad and seafood demand, lower soup demand). Re-classify all items and update the quadrant chart. Add 3 new seasonal specials to the menu list. Redeploy.