• PRECISION
  • CONTEXT
  • HYBRID
  • HUMAN

How to perform hybrid search queries with your AI

Connect Azure AI Search to your agent to combine keyword precision with semantic understanding. You get better results by hitting both text and meaning at once.

Ask AI about this page

Short answer

How do I run hybrid searches with my AI?

Your AI uses Azure AI Search to run both BM25 full-text queries and vector similarity searches at the same time. This approach catches exact keyword matches and conceptual matches in your document collections. You'll see the specific results below.

Search outcomes

Where your results land.

The AI sorts through your data to find the best matches.

PRECISION

Exact keyword hits

The AI uses BM25 logic to find specific terms or serial numbers. You get results that match your exact phrasing.

CONTEXT

Semantic matches

The AI finds documents that mean the same thing as your prompt even if the words differ. This captures intent rather than just spelling.

HYBRID

Combined results

The AI merges keyword and vector scores into a single ranked list. You get the best of both worlds in one response.

HUMAN

Review and refine

You look over the retrieved documents to ensure the context is right. You decide if the search needs more specific keywords.

The workflow

What your AI does when a query arrives.

The AI translates your natural language into a multi-layered search request.

  1. Analyze intent

    Your agent breaks down your question to identify both specific terms and the underlying concept.

    search_documents

  2. Execute vector search

    The AI converts your query into a vector to find conceptually similar documents in the collection.

    vector_search

  3. Execute keyword search

    The AI runs a traditional text search to catch specific identifiers or exact word matches.

    search_documents

  4. Rank and merge

    The system combines the scores from both methods to present the most relevant documents first.

    search_documents

Try it

Copy these to start.

Use these prompts to test the hybrid search capability.

Starting points

Swap the bracketed text for your actual document names or specific search terms.

Azure AI Search Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_UGnljcexY2MC0gUzqrUAvXNWXI0LuEzLwo7dKO1k/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Azure AI Search capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "azure-ai-search-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_UGnljcexY2MC0gUzqrUAvXNWXI0LuEzLwo7dKO1k/mcp"
    }
  }
}
Copy into chat04

  • Find all documents in the [Project Alpha] folder that mention 'safety protocols'.

  • Search for information regarding [Employee Handbook] about remote work policies.

  • Using the [Technical Manuals] collection, find documents related to 'thermal regulation' even if they don't use that exact phrase.

  • Look through [Client Contracts] for any mention of 'termination clauses' or 'notice periods'.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Azure AI Search once, then ask.

Just provide your connection details to link your index. Your credentials stay encrypted, and you can start querying immediately.

Connect Azure AI Search to your AI

FAQ

How hybrid search behaves

  • 01

    Can the AI delete documents from my Azure index?

    No. The tools only allow for searching and retrieving data, not modifying or deleting it.

  • 02

    Does the AI create new vector embeddings for me?

    No. This MCP uses the existing index and embeddings you have already configured in Azure.

  • 03

    What is the difference between the two search types?

    Vector search finds things based on meaning, while the text search finds things based on exact character matches.

  • 04

    Can I use this to upload new files?

    No. This connector is strictly for querying your existing document collections.

  • 05

    Will the AI remember my search history?

    The AI client you use handles your conversation history, not the connector itself.

  • More questions about Azure AI Search? The Connector page answers them. See everything the Azure AI Search Connector can do

Connect Azure AI Search to Claude, Cursor, ChatGPT & more