PHPackages navindbhudiya/module-product-recommendation - PHPackages - PHPackages [Skip to content](#main-content)[PHPackages](/)[Directory](/)[Categories](/categories)[Trending](/trending)[Leaderboard](/leaderboard)[Changelog](/changelog)[Analyze](/analyze)[Collections](/collections)[Log in](/login)[Sign up](/register) 1. [Directory](/) 2. / 3. [Utility & Helpers](/categories/utility) 4. / 5. navindbhudiya/module-product-recommendation ActiveMagento2-module[Utility & Helpers](/categories/utility) navindbhudiya/module-product-recommendation =========================================== AI-Powered Product Recommendations using ChromaDB Vector Database for Magento 2 1.0.0(8mo ago)102MITPHPPHP ^8.1 || ^8.2 || ^8.3CI failing Since Dec 12Pushed 1mo agoCompare [ Source](https://github.com/navindbhudiya-code/magento2-ai-product-recommendations)[ Packagist](https://packagist.org/packages/navindbhudiya/module-product-recommendation)[ Docs](https://github.com/navindbhudiya-aws/magento2-ai-product-recommendations)[ GitHub Sponsors](https://github.com/sponsors/navindbhudiya)[ RSS](/packages/navindbhudiya-module-product-recommendation/feed)WikiDiscussions main Synced 1w ago READMEChangelogDependencies (14)Versions (2)Used By (0) NavinDBhudiya\_ProductRecommendation ==================================== [](#navindbhudiya_productrecommendation) [![CI](https://github.com/navindbhudiya/magento2-product-recommendation/actions/workflows/ci.yml/badge.svg)](https://github.com/navindbhudiya/magento2-product-recommendation/actions/workflows/ci.yml) AI-powered product recommendations for Magento 2 — semantic *related*, *cross-sell*, and *up-sell* suggestions driven by vector embeddings, with optional LLM re-ranking and behaviour-based personalization. > This README describes **only what the code does today**. Planned and partially-built features live in [ROADMAP.md](ROADMAP.md); an honest feature-by-feature breakdown is in [dev/demo/AUDIT.md](dev/demo/AUDIT.md). Status ------ [](#status) Productization phases 0–4 are implemented and **unit-tested** (128 unit + 3 integration tests, phpcs-clean on all new code). The pluggable vector store and the never-empty fallback are wired into the live serving path (`RecommendationService`). Items that require a running stack — admin analytics dashboard UI + JS beacon, GDPR export/erase framework wiring, and all live-environment gate evidence (Luma indexing, hit-rate, CI-green-on-PR) — are listed in [ROADMAP.md](ROADMAP.md). What it does ------------ [](#what-it-does) - **Semantic related / cross-sell / up-sell** via nearest-neighbour search over product embeddings, injected into the native Magento blocks through plugins. - **Pluggable vector store** behind `Api/VectorStoreInterface`: - **ChromaDB** (existing). - **Search Engine** — your store's OpenSearch/Elasticsearch via k-NN (**no extra infra**). - **Pluggable embeddings** behind `Api/EmbeddingProviderInterface`: - **Hosted API** (OpenAI-compatible, e.g. `text-embedding-3-small`) — recommended default. - **ChromaDB embedding-service** (self-hosted Python, `all-MiniLM-L6-v2`). - **Never-empty block**: a fallback chain (primary → same-category → Magento native) keeps the slot populated even when the AI backend is down (`Service/Fallback/FallbackSelector`). - **Personalized recommendations** (browsing / purchase / wishlist / "Just for you") with REST + GraphQL APIs. - **Optional LLM re-ranking** (Claude / OpenAI) — **off by default**. Requirements ------------ [](#requirements) - Magento **2.4.6–2.4.8**, PHP **8.1–8.3**. - A vector store: either OpenSearch/Elasticsearch (already required by Magento) **or** ChromaDB. - For hosted embeddings: an API key (OpenAI/Voyage/compatible). Install in 3 minutes (no extra infra) ------------------------------------- [](#install-in-3-minutes-no-extra-infra) Uses your existing OpenSearch + a hosted embeddings API — no ChromaDB, no Python container. ``` # 1. Add the module (or copy into app/code/NavinDBhudiya/ProductRecommendation) bin/magento module:enable NavinDBhudiya_ProductRecommendation bin/magento setup:upgrade bin/magento setup:di:compile # 2. Configure (Stores > Configuration > NavinDBhudiya > AI Product Recommendation): # - Embedding > Embedding Provider = "Hosted API"; set API key + model (text-embedding-3-small) # - Vector Store > Backend = "Search Engine (OpenSearch/Elasticsearch k-NN)" # (set host/port if not the defaults opensearch:9200) # 3. Index + verify bin/magento recommendation:index bin/magento recommendation:health # expect green + X/Y coverage ``` Prefer ChromaDB? Set Vector Store = ChromaDB and Embedding Provider = ChromaDB, then run the embedding-service container (see `docker/`) — see [CLAUDE.md](CLAUDE.md) for the ChromaDB path. CLI commands ------------ [](#cli-commands) All commands use the `recommendation:*` namespace. CommandPurpose`recommendation:health`Ping embedding provider + vector store; show index coverage.`recommendation:index`Embed and store the catalog.`recommendation:test`Test the ChromaDB / embedding-service connection.`recommendation:similar `Show similar products for a product (or `--query`).`recommendation:clear`Clear the vector collection.`recommendation:personalized`Personalized recommendations for a customer.`recommendation:trending:refresh`Refresh the trending table.`recommendation:refresh-profiles`Refresh customer behaviour profiles.`recommendation:demo:baseline`Measure hit-rate@10 + latency over the ground-truth pairs.Architecture ------------ [](#architecture) ``` Product save / recommendation:index │ ▼ EmbeddingProviderInterface ──► VectorStoreInterface (ChromaDB | Search Engine k-NN) (Hosted API | ChromaDB) ▲ │ │ ▼ PDP / cart blocks ──► RecommendationService ──► nearest neighbours │ ├─ optional LLM re-rank (off by default) └─ FallbackSelector: primary → same-category → native ``` - `Api/VectorStoreInterface` — `upsert / query / delete / count / ping`. Implementations: `Service/VectorStore/ChromaVectorStore`, `Service/VectorStore/SearchEngineVectorStore`(selected by config via `VectorStoreFactory`). - `Api/EmbeddingProviderInterface` — implementations: `ApiEmbeddingProvider`, `ChromaDBEmbeddingProvider` (selected via `EmbeddingProviderFactory`). - REST: `etc/webapi.xml`. GraphQL: `etc/schema.graphqls`. APIs ---- [](#apis) - **REST** — e.g. `GET /V1/recommendation/personalized/justforyou`. - **GraphQL** — `personalizedRecommendations(type: JUST_FOR_YOU, limit: 8) { items { sku name } }`. Testing ------- [](#testing) No Warden required: ``` bash dev/demo/run-tests.sh # unit + integration + phpcs bash dev/demo/run-tests.sh --unit composer test # if composer is available ``` See [CLAUDE.md](CLAUDE.md) for the PHPUnit-10 / Magento classmap caveat and the test launcher. License ------- [](#license) MIT. Author: Navin Bhudiya. ### Health Score 37 — LowBetter than 81% of packages Maintenance77 Regular maintenance activity Popularity3 Limited adoption so far Community10 Small or concentrated contributor base Maturity50 Maturing project, gaining track record Bus Factor1 Top contributor holds 96.2% of commits — single point of failure How is this calculated?**Maintenance (25%)** — Last commit recency, latest release date, and issue-to-star ratio. Uses a 2-year decay window. **Popularity (30%)** — Total and monthly downloads, GitHub stars, and forks. Logarithmic scaling prevents top-heavy scores. **Community (15%)** — Contributors, dependents, forks, watchers, and maintainers. Measures real ecosystem engagement. **Maturity (30%)** — Project age, version count, PHP version support, and release stability. ### Release Activity Cadence Unknown Total 1 Last Release 249d ago ### Community Maintainers ![](https://www.gravatar.com/avatar/18b8c30f0537efcf8a0ce7b3862b904922f7a962509115577cba8761463cc4af?d=identicon)[navindbhudiya](/maintainers/navindbhudiya) --- Top Contributors [![navindbhudiya-code](https://avatars.githubusercontent.com/u/246432367?v=4)](https://github.com/navindbhudiya-code "navindbhudiya-code (25 commits)")[![navinmagento](https://avatars.githubusercontent.com/u/49339423?v=4)](https://github.com/navinmagento "navinmagento (1 commits)") --- Tags aiopenaimachine learningclaudellmmagento2magento 2embeddingschromadbvector-databasepersonalizationrelated productscross-sellproduct recommendationup-sell ### Code Quality TestsPHPUnit Static AnalysisPHPStan Code StylePHP\_CodeSniffer Type Coverage Yes ### Embed Badge ![Health badge](/badges/navindbhudiya-module-product-recommendation/health.svg) ``` [![Health](https://phpackages.com/badges/navindbhudiya-module-product-recommendation/health.svg)](https://phpackages.com/packages/navindbhudiya-module-product-recommendation) ``` ### Alternatives [mollie/magento2 Mollie Payment Module for Magento 2 1142.0M17](/packages/mollie-magento2)[run-as-root/magento2-prometheus-exporter Magento2 Prometheus Exporter 69362.0k](/packages/run-as-root-magento2-prometheus-exporter)[buckaroo/magento2 Buckaroo Magento 2 extension 32426.0k8](/packages/buckaroo-magento2)[dotdigital/dotdigital-magento2-extension Dotdigital for Magento 2 50406.2k23](/packages/dotdigital-dotdigital-magento2-extension)[loki/magento2-components Core module for defining Alpine.js components with advanced AJAX features 1015.1k29](/packages/loki-magento2-components)[yireo/magento2-emailtester2 Preview transactional emails and test send them in your backend 34441.8k](/packages/yireo-magento2-emailtester2) PHPackages © 2026 [Directory](/)[Categories](/categories)[Trending](/trending)[Changelog](/changelog)[Analyze](/analyze)