home / mcp / typesense mcp server

Typesense MCP Server

MCP server for Typesense

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "avarant-typesense-mcp-server": {
      "command": "uv",
      "args": [
        "--directory",
        "~/typesense-mcp-server",
        "run",
        "mcp",
        "run",
        "main.py"
      ]
    }
  }
}

You have an MCP server that bridges Typesense with the Model Context Protocol, enabling you to manage collections, documents, and searches via MCP-compatible clients. This server exposes a set of tools to health-check, describe, index, and query your Typesense data, all through standardized MCP commands you can integrate into your apps and workflows.

How to use

You use this MCP server by calling its available MCP tools through an MCP client. The server exposes capabilities to check health, list and describe collections, create and delete collections, and to index, update, or remove documents. You can perform searches and vector similarity searches against your configured Typesense collections. Think of it as a bridge that lets your MCP client issue high-level data operations against Typesense without talking to the Typesense API directly.

How to install

Prerequisites you need before installing include a runtime environment for the server and the ability to run the MCP runtime you choose.

Manual installation steps you should follow exactly as written below to set up the server locally.

# Prerequisite: install uv on your system (for local runtime)
# Mac users can install via Homebrew
brew install uv

# Clone the MCP server package locally
git clone [email protected]:avarant/typesense-mcp-server.git ~/typesense-mcp-server

# Update your Cursor MCP config to use this server
# Add the following to your .cursor/mcp.json under mcpServers
``````json
{
  "mcpServers": {
    "typesense": {
      "command": "uv",
      "args": ["--directory", "~/typesense-mcp-server", "run", "mcp", "run", "main.py"],
      "env": {
        "TYPESENSE_HOST": "",
        "TYPESENSE_PORT": "",
        "TYPESENSE_PROTOCOL": "",
        "TYPESENSE_API_KEY": ""
      }
    }
  }
}
```

Additional setup guidance

If you prefer an automated installation path, you can install the MCP server through your MCP client’s installer workflow (if available for your environment). The manual steps above provide a local development flow that you can adapt for production by wiring your environment variables and persistent storage as needed.

Available tools

check_typesense_health

Checks the health status of the configured Typesense server

list_collections

Retrieves a list of all collections in the Typesense server

describe_collection

Retrieves the schema and metadata for a specific collection

export_collection

Exports all documents from a specific collection

create_collection

Creates a new collection with the provided schema

delete_collection

Deletes a specific collection

truncate_collection

Truncates a collection by deleting all documents but keeping the schema

create_document

Creates a single new document in a specific collection

upsert_document

Upserts (creates or updates) a single document in a specific collection

index_multiple_documents

Indexes (creates, upserts, or updates) multiple documents in a batch

delete_document

Deletes a single document by its ID from a specific collection

import_documents_from_csv

Imports documents from CSV data into a collection

search

Performs a keyword search on a specific collection

vector_search

Performs a vector similarity search on a specific collection