azure-search-documents-ts

What problem does it solve?

This Skill gives clear, practical TypeScript patterns and operational guidance for building and managing Azure AI Search indexes, ingesting embeddings, and implementing vector, hybrid, and semantic search so teams can deliver relevant retrieval and QA features faster and with fewer integration errors.

Core Features & Use Cases

• Index creation & management: Examples for creating vector-enabled indexes, semantic configurations, and HNSW algorithm settings.
• Document ingestion & batching: Patterns for uploading, merging, and deleting documents including embeddings and incremental updates.
• Retrieval modes: Ready-to-use approaches for vector search, hybrid (text + vector) search, semantic ranking with captions and answers, autocomplete, facets, and filtering.
• Use cases: Building knowledge bases, product search with embeddings, semantic Q&A over document corpora, and hybrid recommender systems.

Quick Start

Use the azure-search-documents-ts skill to create a vector-capable index, upload documents with embeddings, and run a hybrid semantic plus vector search against your AZURE_SEARCH_ENDPOINT and index.