Tukify — AI Shopping Assistant for WooCommerce

Tukify adds an AI shopping assistant to your WooCommerce store. A shopper types in plain language (”a gift for my dad, he likes wine and hiking”) and Tukify understands the intent, finds matching products with semantic search, and replies conversationally with product cards — image, price, stock, and add-to-cart — right inside a chat interface, without the shopper leaving the page.

Tukify is bring-your-own-key: you connect your own AI provider — Google Gemini, OpenAI (ChatGPT), Anthropic (Claude), or xAI (Grok) — and Tukify never pays for inference. All AI requests are made server-side in PHP — your API keys are never exposed to the browser. Your product embeddings are stored in your own site database, so your catalog data stays on your server.

Key features

• Semantic product search — matches by meaning, not just keywords (”warm clothes for winter” surfaces jackets and sweaters).
• Conversational RAG chat — grounded answers that only ever recommend real, in-stock products; it never invents items.
• Clarifying questions — asks one short follow-up (with tappable quick replies) when a request is too vague, instead of dumping mediocre results.
• Natural-language filters — understands price, colour, size, brand, stock, category and sorting (”cheapest wireless headphones under $200”) and applies them as real WooCommerce queries.
• Comparison mode — side-by-side comparison of 2–3 products with a one-line trade-off summary.
• Policy / FAQ answers — answers shipping, returns and warranty questions from your own pages and a custom Q&A knowledge base.
• Cart-aware upsell — suggests genuinely complementary items based on the cart.
• Exit-intent re-engagement — a helpful, dismissible prompt when a shopper is about to leave (once per visit, with a cooldown).
• Visual search — a shopper can upload an image and Tukify finds similar products in your catalog.
• Background indexing — embeds your catalog in the background via Action Scheduler; only re-embeds changed products.
• Presentation surfaces — a global floating chat widget plus three Elementor widgets (Chat, Search, Recommendations), each rendered in a Shadow DOM so your theme’s CSS can’t break them.
• Analytics dashboard — top queries, zero-result queries, click-through and chat-to-sale.

Privacy and data

All AI calls are made server-side; your API key is never sent to the browser. Product embeddings are stored in your own site’s database. Tukify only contacts an external service (Google Gemini) when you have configured an API key and a shopper or the store uses a feature that needs it. See the External services section below for exactly what is sent and when.

Requirements

• WooCommerce (active).
• An API key for at least one supported AI provider: Google Gemini, OpenAI, Anthropic (Claude), or xAI (Grok). Gemini offers a free tier that is enough for development and small stores.
• If you pick Claude or Grok for chat, you also need a Gemini or OpenAI key for embeddings (those two providers have no embeddings endpoint).
• PHP 7.4+.
• Elementor is optional (only needed for the Elementor widgets).

Which provider does what

• Google Gemini — chat, embeddings, vision (image search). Default.
• OpenAI (ChatGPT) — chat, embeddings, vision.
• Anthropic (Claude) — chat, vision. No embeddings — use Gemini or OpenAI as the embedding provider.
• xAI (Grok) — chat, vision. No embeddings — use Gemini or OpenAI as the embedding provider.

You choose a chat provider and an embedding provider independently in Settings. Retrieval (RAG) always uses the embedding provider. If you change the embedding provider or model, Tukify prompts you to reindex, because embeddings from different models are not compatible and are never mixed in the same index.

External services

Tukify connects to the AI provider(s) you configure to power semantic search, conversational answers, natural-language understanding, and image-based (visual) search. A provider is only contacted when you have entered its API key and selected it, and only for the feature that needs it. All requests are made from your server (PHP); your API keys are never exposed to the browser, and Tukify does not send data to any other third party.

What data is sent, and when:

• Catalog indexing — when you index or re-index your products, the text of each product (title, short description, categories and key attributes) is sent to your embedding provider to generate numeric embeddings. This runs in the background and only for products that have changed.
• Knowledge base indexing — if you enable the policy/FAQ knowledge base, the content of the pages you select and any custom Q&A you enter is sent to your embedding provider.
• Search and chat — when a shopper searches or chats, their message, a short recent conversation history, and the retrieved product context are sent to your embedding provider (to embed the query) and your chat provider (to generate a grounded reply).
• Visual search — if a shopper uploads an image, that image is sent to a vision-capable provider to identify the product type before matching your catalog.

Provider terms and privacy policies (review the ones you use):

• Google Gemini — https://ai.google.dev/terms · https://policies.google.com/privacy
• OpenAI — https://openai.com/policies/terms-of-use · https://openai.com/policies/privacy-policy
• Anthropic (Claude) — https://www.anthropic.com/legal/consumer-terms · https://www.anthropic.com/legal/privacy
• xAI (Grok) — https://x.ai/legal/terms-of-service · https://x.ai/legal/privacy-policy

Note: some providers may use free-tier inputs to improve their models. If you handle real customer data, use a paid plan/key (typically excluded from training) and review that provider’s terms.