PlantDex

PlantDex is a plant discovery site built as a UI and index over the Trefle plant API. Search a familiar common name (or a scientific one), open rich species profiles, explore related plants by taxonomy and traits, compare a few side by side, and skim a short AI overview — without accounts or personal collections.

The product goal is botanical context, not identification from photos or garden tracking. Searching “blueberry,” for example, should help you see related Vaccinium species and Ericaceae relatives, not just a single shopping card.

See DesignDoc.md for full product scope, API surface, and architecture.

What you can do

• Search plants by common or scientific name, with paginated “load more” results
• Profiles with taxonomy, traits, distribution chips, and a globe for native + introduced ranges
• Related discovery by genus, family, distribution, edible part, growth habit, fruit color, and more
• Compare 2–4 plants in a local tray / compare page (browser storage only)
• AI overviews (summary text) on search results and profiles — imperfect; treat as a starting point
• Report / correct Trefle data through the app (API token stays on the server)

Structure

• frontend/ — Next.js app (Vercel)
• src/ — Python Lambda handlers + Trefle similarity helpers
• infrastructure/ — AWS CDK (API Gateway, Lambda, DynamoDB, Secrets Manager)
• notebooks/ / tests/ — exploration and notebooks

Frontend

cd frontend
cp .env.example .env.local
# set NEXT_PUBLIC_PLANTDEX_API_URL and OPENAI_API_KEY (server-only)
npm install
npm run dev

Deploy on Vercel with:

• Root Directory: frontend
• Production branch: main
• Env: NEXT_PUBLIC_PLANTDEX_API_URL and OPENAI_API_KEY (Production + Preview; never NEXT_PUBLIC_ for the OpenAI key)

Backend

aws sso login --profile plantdex
cd infrastructure
source .venv/bin/activate
AWS_PROFILE=plantdex cdk deploy

The Lambda stack proxies Trefle reads/writes, caches responses in DynamoDB, and keeps the Trefle token in Secrets Manager.