PlantFolio 🌿
Where your plants tell their story.
A mobile-responsive web platform where plant lovers build a profile for every plant they own, track its health, trade plants with others, and diagnose problems from a photo. Built for the Plant Care & Exchange Community coursework brief — B.Sc. (Hons) Computer Science, Wayamba University of Sri Lanka — and extended with two machine-learning features: plant species recognition and plant disease detection.
Status — 4 Oct 2026: planning and design are complete (proposal, 25-page wireframes, clickable Figma prototype, this repo).
Next: frontend-first build starts Monday 5 Oct 2026 — all 15 pages with mock data in weeks 1–7, then PHP + MySQL, then ML.
See docs/ROADMAP.md.
Features
Core (coursework brief) — secure sign-up and login · plant profiles with species care information · named collections (public or private) · health tracking with a dated plant story · care reminders with a due/overdue dashboard · search and discovery (plants and @usernames) · exchange marketplace with interest requests and contact reveal · trader ratings · responsive design for phone, tablet and desktop.
ML enhancements — ✦ Plant Identifier: top-3 species from a photo, applied to a new plant in one click · ✦ Health Check: a CNN predicts leaf disease with a confidence score and an action plan. Both are helpers: below 60 % confidence they say "Not sure", and nothing is saved without the user's OK.
Design
| |
|---|
| Clickable prototype | PlantFolio on Figma — journeys A–D, 32 desktop screens, 4 mobile screens, design-system page |
| Wireframes & UX structure | docs/PlantFolio_Wireframes_UX_Structure.pdf — 15 pages, design system, states, build handoff |
| Project proposal | docs/PlantFolio — Project Proposal.pdf — scope, architecture, database, ML plan, risks |
To present the prototype: open the 🌿 Prototype page → Present → pick a flow (D · First visit, A · Add an unknown plant, B · Diagnose a sick plant, C · Swap a plant, Mobile · 375 px). P07, P09, P12, P14 and P15 are not in the prototype yet.
Tech stack
| Layer | Technology |
|---|
| Front end | HTML5, hand-written CSS with design tokens (no CSS framework), vanilla JavaScript · mobile-first from 375 px · Lora + Poppins, Lucide icons |
| Back end | Plain PHP 8+ with PDO (no framework) |
| Database | MySQL (InnoDB, utf8mb4) |
| ML service | Python 3.10+, Flask, TensorFlow/Keras — models trained on Google Colab, called from PHP over REST |
| Local server | XAMPP (Apache + PHP + MySQL) |
| Tools | VS Code, Git + GitHub (main ← develop ← feature/*), Figma |
How we're building it
We build frontend first: every page is made to match its wireframe using mock data shaped like real database rows, then wired to PHP and MySQL, then the ML service is added. The three quality gates from the proposal stay the same — they just land later.
| Stage | Weeks | Dates | Outcome |
|---|
| 0 · Planning & design | 0 | to 4 Oct | ✅ Proposal, wireframes, Figma prototype, repo |
| F · Frontend with mock data | 1–7 | 5 Oct – 22 Nov | All 15 pages and shared states, responsive, no backend → UI freeze |
| B · Backend & database | 8–10 | 23 Nov – 13 Dec | Pages wired to PHP + MySQL → Gate 1 (wk 9) · Gate 2: coursework complete (wk 10, tag v1.0-core) |
| M · ML features | 11–14 | 14 Dec – 10 Jan | Flask service, disease + species models, identifier and health check → Gate 3 |
| D · Testing & delivery | 15–16 | 11 – 24 Jan | Security review, usability tests, report, slides, release v1.0 |
Dates assume the submission is around late January 2027 — they will be re-baselined once the official date is confirmed (task PLN-04). Full plan with task IDs: docs/ROADMAP.md.
Folder structure
plantfolio/
├── public/ # Web root — one PHP file per page (index.php … settings.php)
│ ├── api/ # small JSON endpoints for fetch() calls
│ ├── assets/ # css/ (variables, components, layout) · js/ · img/
│ └── uploads/ # user photos (git-ignored)
├── includes/ # mock-data.php (frontend stage), header/footer, helpers, db, auth, upload, ml-client
├── partials/ # plant-card.php, listing-card.php and other reusable pieces
├── config/ # config.example.php (config.php is git-ignored)
├── database/ # schema.sql, seed.sql, ERD, species care data
├── ml-service/ # Flask API, notebooks/, models/ (git-ignored)
├── docs/ # Proposal, wireframes, roadmap, conventions, report drafts
├── .github/ # Pull-request template
└── CLAUDE.md # Guide for Claude Code (stack, rules, design tokens)
Run it locally
- Install XAMPP (PHP 8+), Git and VS Code.
- Clone into the XAMPP web folder and switch to
develop:
cd C:/xampp/htdocs # macOS: /Applications/XAMPP/htdocs
git clone https://github.com/rSith/plantfolio.git
cd plantfolio
git checkout develop
(develop is created once by the leader in task PLN-09; until then use main.)
- Create your local config:
cp config/config.example.php config/config.php, then edit it.
- Start Apache in the XAMPP Control Panel and open http://localhost/plantfolio/public/.
- From week 8 (stage B): start MySQL, create a database called
plantfolio in phpMyAdmin and import database/schema.sql, then database/seed.sql.
- From week 11 (stage M): start the ML service — see
ml-service/README.md. The site still works without it.
During stage F no database is needed — pages read from includes/mock-data.php.
This is the beginning of the README. Read the rest at the original. Read the full README ↗
