Laravel Scout

Agent Workflow (MANDATORY)

Before ANY implementation, use TeamCreate to spawn 3 agents:

1. fuse-ai-pilot:explore-codebase - Analyze existing model and search patterns
2. fuse-ai-pilot:research-expert - Verify Scout docs via Context7
3. mcp__context7__query-docs - Check search and indexing patterns

After implementation, run fuse-ai-pilot:sniper for validation.

---

Overview

Component	Purpose
Searchable Trait	Makes Eloquent models searchable
Search Drivers	Meilisearch, Algolia, database, collection
Indexing	Automatic sync on model changes
Search Builder	Fluent search API with filters

---

Decision Guide: Search Driver

Which driver?
├── Production (recommended) → Meilisearch (fast, self-hosted, free)
├── Managed service → Algolia (hosted, pay per search)
├── Small dataset → database (no extra infra)
└── Testing → collection (in-memory, no engine)

---

Quick Setup

composer require laravel/scout
composer require meilisearch/meilisearch-php http-interop/http-factory-guzzle

SCOUT_DRIVER=meilisearch
MEILISEARCH_HOST=http://127.0.0.1:7700
MEILISEARCH_KEY=masterKey

$results = Article::search('laravel tutorial')->paginate(15);

---

Critical Rules

1. Use toSearchableArray() to control indexed data
2. Queue indexing with SCOUT_QUEUE=true for performance
3. Use searchable() for bulk import after setup
4. Pause indexing during seeders with Scout::withoutSyncing()

---

Reference Guide

Need	Reference
Searchable trait, indexing, conditions	searchable.md
Driver setup, Meilisearch, Algolia	drivers.md

---

Best Practices

DO

• Use Meilisearch for production (fast, typo-tolerant)
• Queue indexing operations (SCOUT_QUEUE=true)
• Limit indexed fields with toSearchableArray()

DON'T

• Index sensitive data (passwords, tokens)
• Forget to import existing records after setup
• Use collection driver in production

---

Laravel 13 Notes

Vector search natif pgvector

Pour la recherche sémantique (embeddings) sur PostgreSQL, Laravel 13 expose Schema::ensureVectorExtensionExists() et whereVectorSimilarTo() via la skill dédiée [[laravel-vector-search]]. Scout reste pertinent pour le full-text (Meilisearch/Algolia) ; pour la similarité vectorielle, utiliser pgvector directement sans driver Scout.

// Hybride : Scout pour full-text, pgvector pour similarité
$keyword = Post::search($query)->get();
$semantic = Post::whereVectorSimilarTo('embedding', $embedding, limit: 10)->get();