🚀 Live Demo · Portfolio & downloads · Case study PDF
A language for what you smell. A connected classifier, digital and print menu studio, PDF label maker, education library, and cultivator creative studio by CannaCre8ive. A fingerprint ID and proportional band carry a reviewed flower batch through every tool.
-
Upload & create: read PDF, CSV, PNG, JPG, or WebP terpene reports on the device; review source rows and units; reconcile the total; add a flower photograph; generate social and chemovar assets.
-
Workflows: practical farm, buyer, and retail-staff journeys. Buyer search supports aroma score ranking, measured terpene minimums, reviewed-upload filtering, CSV shortlist export, and shortlist-to-print handoff.
-
Classifier: the full 38-terpene model, flower/live-resin presets, detailed and spectrum fingerprints, four-panel comparison, CSV export, and save/update to the shared catalog.
-
Print menu: flower only, staff picks, deals, sorting, editable headers, portrait/landscape pagination, and ten profile plus 38 terpene shelf cards.
-
Labels: Avery 6464 PDFs; product selection, copy order, start position, automatic clean canvas preview of the actual PDF, and download.
-
Profile assets: ten layered illustrative blends as 1080 × 1080 PNG cards and editable SVG fingerprints; a historical Kush Mints walkthrough and interactive concentration/weighting explanation.
-
Digital menu: flower only, aroma filters, sorting, product details, contextual learning, staff picks, demo sales, and local catalog editing.
-
Education: nine supplied teaching pieces, individual/full print views, and an interactive preference quiz that links into retail browsing.
-
Social studio: six customizable PNG formats, three historical sample panels, detailed chemovar cards, and shared profile colors.
-
Portfolio: a case study, real screenshots, print primer, social examples, and a complete downloadable asset kit.
npm ci
npm run devOpen the local address shown by Vite. npm run check runs source-preservation, classification, and CSV regression checks followed by a production build. npm run preview serves the production build. Use Node 22.12+ or a newer supported LTS. No environment variables or credentials required.
| Folder | Purpose |
|---|---|
source/ |
Exact original HTML/JSX plus SHA-256 manifest |
src/data/ |
Shared profiles, terpene reference, sample panels, and catalog seeds |
src/lib/ |
Preserved classification math, CSV handling, local catalog state |
src/modules/ |
Six adapted tools and profile asset gallery |
portfolio/ |
Editable case study, asset guide, ready-to-use captions |
public/assets/ |
Downloadable PDF, PNG, and ZIP portfolio deliverables |
documentation/assets/ |
Real app screenshots |
The React shell lazy-loads all six tool views. The social canvas has its own HTML entry to preserve its styling while importing shared data modules. Browser catalog changes stay on this device; no server stores or publishes them. Export a reviewed record as JSON to preserve its source rows and photograph. Menu CSV preserves catalog fields and panels but not the uploaded photo or review audit.
This is a portfolio demonstration, not live inventory or a production retail system. Pricing, staff attributions, and most product values are illustrative. Historical sample panels were transcribed in the supplied sources; original laboratory PDFs were not included or independently verified. Relative aroma-model scores are not measured terpene concentrations or effect predictions. The shared classifier models 38 terpenes; unmodeled entries stay in the catalog and are disclosed by the classifier. Raw-score ordering resolves ties before displayed percentages are rounded. Legacy education copy requires editorial review before retail use.
The application does not upload imported catalogs. It has no authentication, checkout, POS integration, or cloud sync. The source files remain intact. See SOURCE-MAP, ARCHITECTURE, and ROADMAP.
Classify a product and save it to the catalog. Open Digital menu to browse it, Print menu to format flower, or Labels to select products and generate a PDF. Print menus at US Letter, 100% scale, with browser headers/footers disabled and background graphics enabled. Print label PDFs at Actual size / 100%. Test on plain paper first; physical printer alignment has not been verified. Export menu CSV for pricing and panel backup; use reviewed-record JSON for the upload audit and photo.
3.1.0: layered profile fingerprints and proportional full-spectrum bands, shared with Education and print references; historical Kush Mints example; interactive explanation of why the most abundant terpene need not determine the leading profile. Reference compositions are explicitly illustrative, and model weights are distinguished from measured smell intensity.
3.0.0: fingerprint-led reference assets; on-device PDF/CSV/image report import and review; optional flower photos; uploaded-flower social and chemovar exports; flower-only menus and labels; live clean label previews; farm/buyer/staff workflows and buyer shortlist search.
2.0.0: shared 38-terpene classifier; measured print pagination; connected PDF labels; ten-profile asset library; full panel preservation in catalog CSV; 900-product browser/PDF stress verification.
1.0.0: unified navigation and profile library; connected historical samples; persistent local catalog; CSV field preservation and validation; responsive layout; six native-size social exports; case study and portfolio asset pack; complete social metadata and stable demo link. See CHANGELOG.
Product requirements · Design system · User flows · Testing · Contributing · Asset guide
Source artwork and supplied content remain subject to their owners' rights. No additional open-source license is granted for the supplied creative material.