Azure AI Search Index Designer

Visual schema builder for Azure AI Search indexes directly in VS Code. Define fields, analyzers, scoring profiles, semantic configurations, and vector search dimensions in a YAML/JSON editor with validation, preview, and one-click deploy. Includes a built-in query tester.

Features

• Visual Schema Designer — Interactive webview to add/edit/remove fields, toggle attributes (searchable, filterable, sortable, facetable, key), and configure vector fields with dimensions and profiles.
• YAML & JSON Support — Write index schemas in .search-index.json or .search-index.yaml files.
• Live Validation — Validates against Azure AI Search REST API rules: field types, vector constraints, scoring function parameters, semantic field references, HNSW algorithm ranges, and analyzer cross-references.
• One-Click Deploy — Deploy indexes directly to Azure AI Search via the REST API.
• Query Tester — Interactive search query builder with support for filters, facets, highlights, semantic search, and vector queries.
• CodeLens Actions — Inline links above index files: Open Designer, Validate, Preview, Deploy, Query Tester.

Getting Started

1. Create an Index File

Run AI Search: New Index File from the Command Palette, or create a file ending in .search-index.json:

{
  "name": "products",
  "fields": [
    { "name": "id", "type": "Edm.String", "key": true, "filterable": true },
    { "name": "title", "type": "Edm.String", "searchable": true, "retrievable": true },
    { "name": "description", "type": "Edm.String", "searchable": true, "retrievable": true },
    { "name": "category", "type": "Edm.String", "filterable": true, "facetable": true },
    { "name": "price", "type": "Edm.Double", "filterable": true, "sortable": true, "facetable": true },
    { "name": "embedding", "type": "Collection(Edm.Single)", "searchable": true, "dimensions": 1536, "vectorSearchProfile": "my-profile" }
  ],
  "vectorSearch": {
    "algorithms": [
      { "name": "hnsw-algo", "kind": "hnsw", "hnswParameters": { "m": 4, "efConstruction": 400, "efSearch": 500, "metric": "cosine" } }
    ],
    "profiles": [
      { "name": "my-profile", "algorithm": "hnsw-algo" }
    ]
  },
  "semantic": {
    "configurations": [
      {
        "name": "default",
        "prioritizedFields": {
          "titleField": { "fieldName": "title" },
          "contentFields": [{ "fieldName": "description" }]
        }
      }
    ]
  }
}

2. Configure Settings

Set your Azure AI Search endpoint and API key in VS Code settings:

3. Use CodeLens or Commands

Click the CodeLens actions above your index file, or use the Command Palette:

Validation Rules

The validator checks:

• Field names and types against all 25+ EdmTypes
• Exactly one key field (must be Edm.String)
• Attribute compatibility (searchable only for text, sortable not for collections, etc.)
• Vector fields: dimensions 2–4096, valid vector types, profile references
• HNSW parameters: m (4–10), efConstruction (100–1000), efSearch (100–1000)
• Scoring profile function parameters (magnitude ranges, freshness durations, etc.)
• Semantic configuration field references must exist and be searchable
• Analyzer/tokenizer/filter cross-references against 100+ built-in names
• Duplicate name detection across fields, profiles, algorithms

Supported Field Types

Edm.String, Edm.Int32, Edm.Int64, Edm.Double, Edm.Boolean, Edm.DateTimeOffset, Edm.GeographyPoint, Edm.ComplexType, Edm.Single, Edm.Half, Edm.Int16, Edm.SByte, Edm.Byte, and their Collection() variants.

Requirements

• VS Code 1.85+
• Azure AI Search service with an admin API key

License

MIT