What problem does it solve? It guides engineers through implementing and operating full-text search over product data (agents, topics, messages, files, knowledge bases) across pg_search, pg_like, and Elasticsearch providers while preserving permission boundaries and stable result contracts. ## Core Features & Use Cases - Provider-neutral search architecture: Routes all queries through FtsSearchRepo with deployment-level provider selection, keeping result shapes and permission hydration stable. - Elasticsearch mapping migrations: Manages index generations, schema fingerprints, checkpoints, promotion, rollback, and in-place additive upgrades. - Outbox-based sync and reindexing: Covers durable change capture, claims, retries, dead letters, resumable backfills, and continuous incremental draining. - Use Case: When adding a new searchable entity, follow the cross-layer checklist to update document schemas, mappings, builders, capture triggers, and reindex checkpoints consistently. ## Quick Start Ask the assistant to add a new searchable entity to the full-text search system and update the mappings, builders, and sync infrastructure accordingly.
