• MATCH
  • CONTEXT
  • RANK
  • HUMAN

How to run semantic search queries with your AI

Connect Marqo AI to your agent to turn natural language questions into precise vector lookups. Your AI stops guessing keywords and starts understanding intent.

Ask AI about this page

Short answer

How can I perform semantic searches using my AI?

Your agent uses the tensor_search tool to query Marqo vector indices using natural language. This allows you to find information based on the actual meaning of your query instead of matching specific words. You get relevant context pulled directly into your chat.

Search outcomes

Where your data lands.

What happens when your agent queries the index.

MATCH

Semantic hits

The AI identifies data points that match the intent of your question. It pulls the most relevant vectors from the Marqo index.

CONTEXT

Rich retrieval

Your agent receives the actual content associated with the vector matches. This provides the raw data needed to answer your prompt.

RANK

Relevance scoring

Marqo ranks results by how closely they align with your query. Your AI uses these scores to prioritize the best information.

HUMAN

Manual review

You decide if the retrieved context is sufficient. If the search is too broad, you'll need to refine your natural language phrasing.

The workflow

What your AI does when a query arrives.

The agent translates your plain English into a mathematical search.

  1. Interpret intent

    Your agent reads your question and determines the semantic meaning behind the words.

    tensor_search

  2. Execute lookup

    The agent calls the Marqo index to find vectors that sit near your query in the embedding space.

    tensor_search

  3. Filter results

    The AI evaluates the returned tensors to ensure they meet the specific criteria of your request.

    tensor_search

  4. Synthesize answer

    Your agent takes the retrieved text and formats it into a coherent response for you.

    tensor_search

Try it

Copy these to start.

Use these prompts to trigger your first semantic searches.

Starting points

Replace the bracketed text with your specific index names or search topics.

Marqo AI Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_yWwQAD75WfnpbXY72FMh8SBRNTeZWrryXTJG2Ifa/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 — Marqo AI capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "marqo-ai-vector-search-embeddings-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_yWwQAD75WfnpbXY72FMh8SBRNTeZWrryXTJG2Ifa/mcp"
    }
  }
}
Copy into chat04

  • Search the [index_name] index for documents discussing [topic].

  • Find all entries in my Marqo index that relate to [specific concept].

  • Using the vector index, find information about [subject] that isn't explicitly named.

  • Retrieve the most relevant context from [index_name] regarding [query].

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

Start here

Connect Marqo AI once, then ask.

Enter your Marqo credentials to link your indices. Once connected, your agent can immediately start querying your data.

Connect Marqo AI to your AI

FAQ

How the search behaves

  • 01

    Does the AI search by exact keyword matches?

    No. It uses semantic embeddings to find matches based on meaning and context.

  • 02

    Can the AI delete data from my Marqo index?

    No. The tensor_search tool only allows for reading and retrieving data from your indices.

  • 03

    Can I search across multiple indices at once?

    The AI performs searches against the specific index you target in your prompt.

  • 04

    Will the AI update my vector embeddings automatically?

    No. This connector is for querying existing data, not for managing or updating the index structure.

  • 05

    How does the AI handle vague questions?

    The AI will attempt to find the closest semantic matches, but you may need to be more specific to get precise results.

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

Connect Marqo AI to Claude, Cursor, ChatGPT & more