PHPackages rubyat/laravel-rag - 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. [Search & Filtering](/categories/search) 4. / 5. rubyat/laravel-rag ActiveLibrary[Search & Filtering](/categories/search) rubyat/laravel-rag ================== Add semantic search and Retrieval-Augmented Generation (RAG) to any Laravel app with PostgreSQL + pgvector. v0.1.1(1mo ago)10MITPHPPHP ^8.3 Since Jun 2Pushed 1mo agoCompare [ Source](https://github.com/rubyat/laravel-rag)[ Packagist](https://packagist.org/packages/rubyat/laravel-rag)[ RSS](/packages/rubyat-laravel-rag/feed)WikiDiscussions main Synced 1w ago READMEChangelogDependencies (3)Versions (4)Used By (0) rubyat/laravel-rag ================== [](#rubyatlaravel-rag) [![Latest Version](https://camo.githubusercontent.com/8fdee0d5c09e4f371ffd89b50c215dda983e3934aaecc0a5688216706f70ca60/68747470733a2f2f696d672e736869656c64732e696f2f7061636b61676973742f762f7275627961742f6c61726176656c2d7261672e737667)](https://packagist.org/packages/rubyat/laravel-rag)[![Total Downloads](https://camo.githubusercontent.com/b9aae5536f50ecf0bd533a1237b3e9b47d1a6a1564b143032866db4716f51ba4/68747470733a2f2f696d672e736869656c64732e696f2f7061636b61676973742f64742f7275627961742f6c61726176656c2d7261672e737667)](https://packagist.org/packages/rubyat/laravel-rag)[![License](https://camo.githubusercontent.com/43dcb0b520657f180404af5633e9d33a8eff8a3e8c8a3d2ffc9382eeae6579d1/68747470733a2f2f696d672e736869656c64732e696f2f7061636b61676973742f6c2f7275627961742f6c61726176656c2d7261672e737667)](LICENSE) A production-ready Retrieval-Augmented Generation (RAG) toolkit for Laravel with PostgreSQL + pgvector. Ingest documents, store embeddings, and answer questions grounded **only** in your own content — with source citations. Features -------- [](#features) - 🧩 **Document ingestion** — text is chunked (configurable size/overlap), embedded, and stored. - 🔎 **pgvector vector search** — cosine similarity over an **HNSW** index for high recall at any scale. - 💬 **RAG pipeline** — retrieves the most relevant chunks and asks an LLM to answer from them. - 📎 **Source citations** — every answer returns the chunks it used, with similarity scores. - 🔌 **Driver-based** — swap embedding/chat providers; OpenAI built-in, plus deterministic `fake` drivers for offline dev and tests (no API key, no network). - ⚙️ **Sync or queued ingestion** — inline by default, or dispatch a job for large documents. - 🌐 **HTTP API** — ready-to-use `POST /api/rag/ingest` and `POST /api/rag/ask` endpoints. - ✅ **Tested** — ships with a Pest suite that runs against pgvector using the fake drivers. Requirements ------------ [](#requirements) - PHP **8.3+** - Laravel **11, 12, or 13** - PostgreSQL with the [`pgvector`](https://github.com/pgvector/pgvector) extension (**0.5+** for HNSW) Installation ------------ [](#installation) ``` composer require rubyat/laravel-rag ``` Publish the config and run the migration: ``` php artisan vendor:publish --tag=rag-config php artisan migrate ``` > The migration runs `CREATE EXTENSION IF NOT EXISTS vector`, which needs a database role allowed to create extensions. The [pgvector Docker image](https://hub.docker.com/r/pgvector/pgvector) (`pgvector/pgvector:pg16`) is the quickest way to get it locally. Configure a provider (or keep the offline `fake` drivers — no network needed): ``` OPENAI_API_KEY=sk-... RAG_EMBEDDING_DRIVER=openai RAG_CHAT_DRIVER=openai ``` Quick Start ----------- [](#quick-start) ### In code [](#in-code) ``` use Rubyat\LaravelRag\Ingestion\DocumentIngestor; use Rubyat\LaravelRag\Rag\RagPipeline; // Ingest a document (chunk -> embed -> store in pgvector) app(DocumentIngestor::class)->ingest('handbook.md', $longText); // Ask a question grounded in the ingested chunks $result = app(RagPipeline::class)->ask('How does pgvector search work?'); $result['answer']; // string — grounded in your documents $result['citations']; // array — [{ source, chunk_index, content, score }, ...] ``` ### HTTP API [](#http-api) Ingest: ``` curl -X POST http://localhost:8000/api/rag/ingest \ -H "Content-Type: application/json" \ -d '{"source": "handbook.md", "content": "Postgres pgvector stores embeddings and supports cosine similarity search..."}' # => 201 { "message": "Document ingested.", "source": "handbook.md", "chunks": 1 } ``` Ask: ``` curl -X POST http://localhost:8000/api/rag/ask \ -H "Content-Type: application/json" \ -d '{"question": "How does pgvector search work?", "top_k": 4}' # => 200 # { # "answer": "pgvector stores embeddings and supports cosine similarity search...", # "citations": [ # { "source": "handbook.md", "chunk_index": 0, "content": "...", "score": 0.83 } # ] # } ``` Configuration ------------- [](#configuration) Published to `config/rag.php`. Every option is environment-driven: OptionEnvDefaultDescription`embedding_driver``RAG_EMBEDDING_DRIVER``openai``openai` or `fake``chat_driver``RAG_CHAT_DRIVER``openai``openai` or `fake``dimensions``RAG_DIMENSIONS``1536`Vector size; must match the embedding model`chunk_size``RAG_CHUNK_SIZE``1000`Chunk length in characters`chunk_overlap``RAG_CHUNK_OVERLAP``200`Overlap between chunks`top_k``RAG_TOP_K``4`Chunks retrieved per question`queue_ingestion``RAG_QUEUE_INGESTION``false`Dispatch ingestion to the queue (needs a worker)`register_routes``RAG_REGISTER_ROUTES``true`Auto-register the HTTP routes`route_prefix``RAG_ROUTE_PREFIX``api/rag`Prefix for the package routesThe `fake` drivers are deterministic and require no network access — they power the test suite and let you try the full flow without API keys. Swappable drivers ----------------- [](#swappable-drivers) Embedding and chat providers are resolved from config and implement small contracts, so you can plug in your own: ``` use Rubyat\LaravelRag\Contracts\EmbeddingDriver; use Rubyat\LaravelRag\Contracts\ChatDriver; ``` Built in: `OpenAiEmbeddingDriver` / `OpenAiChatDriver` (HTTP) and `FakeEmbeddingDriver` / `FakeChatDriver` (offline, deterministic). How it works ------------ [](#how-it-works) ``` ingest ask │ │ ▼ ▼ DocumentChunker VectorRetriever │ chunks │ query embedding ▼ ▼ EmbeddingDriver ─── embeddings ──► pgvector documents (HNSW, cosine) │ top-k chunks ▼ RagPipeline (context + LLM) ▼ answer + citations ``` Documents are stored one row per chunk in the `documents` table (`source`, `chunk_index`, `content`, `metadata`, `embedding vector(n)`), indexed with HNSW (`vector_cosine_ops`). Retrieval orders by cosine distance (`embedding :query`) and returns a similarity score in `[0, 1]`. Testing ------- [](#testing) The suite runs against PostgreSQL + pgvector using the deterministic `fake` drivers, so no API keys or network are required: ``` ./vendor/bin/pest ``` License ------- [](#license) The MIT License (MIT). See [LICENSE](LICENSE). ### Health Score 36 — LowBetter than 79% of packages Maintenance90 Actively maintained with recent releases Popularity2 Limited adoption so far Community6 Small or concentrated contributor base Maturity41 Maturing project, gaining track record Bus Factor1 Top contributor holds 100% 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 Every ~0 days Total 2 Last Release 53d ago ### Community Maintainers ![](https://avatars.githubusercontent.com/u/14076245?v=4)[MD Rubyat Islam](/maintainers/rubyat)[@rubyat](https://github.com/rubyat) --- Top Contributors [![rubyat](https://avatars.githubusercontent.com/u/14076245?v=4)](https://github.com/rubyat "rubyat (4 commits)") --- Tags embeddingslaravelopenaipgvectorphpragsemantic-searchvector-searchlaravelopenaiembeddingsragpgvectorvector-searchsemantic-search ### Embed Badge ![Health badge](/badges/rubyat-laravel-rag/health.svg) ``` [![Health](https://phpackages.com/badges/rubyat-laravel-rag/health.svg)](https://phpackages.com/packages/rubyat-laravel-rag) ``` ### Alternatives [psalm/plugin-laravel Psalm plugin for Laravel 3345.3M347](/packages/psalm-plugin-laravel)[laravel/ai The official AI SDK for Laravel. 1.0k3.2M246](/packages/laravel-ai)[api-platform/laravel API Platform support for Laravel 58174.6k17](/packages/api-platform-laravel)[simplestats-io/laravel-client Server-side analytics for Laravel that follows the full funnel from visit to registration to payment, attributed to the channel that drove it. Revenue, MRR, churn and ad-spend profit (ROAS/CAC) per channel. GDPR compliant, ad-blocker proof. 5022.6k](/packages/simplestats-io-laravel-client)[aedart/athenaeum Athenaeum is a mono repository; a collection of various PHP packages 255.2k](/packages/aedart-athenaeum) PHPackages © 2026 [Directory](/)[Categories](/categories)[Trending](/trending)[Changelog](/changelog)[Analyze](/analyze)