Typesense MCP Server

Name: Typesense MCP Server
Author: avarant

@avarant

Repository

About Typesense MCP Server

MCP server for Typesense

#python #mcp #mcp-server

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "typesense-mcp-server": {
      "command": "uv",
      "args": [
        "run",
        "python",
        "main.py"
      ],
      "env": {
        "TYPESENSE_API_KEY": "xyz"
      }
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Typesense MCP Server?

A Model Context Protocol (MCP) server that provides a standardized interface to interact with Typesense, a fast open-source search engine. It is designed for developers using MCP-compatible clients (like Claude Desktop, Cursor, Zed, VS Code) to perform search operations, manage collections, and handle documents in Typesense through natural language commands.

How to use Typesense MCP Server?

Install uv, clone the repository to ~/typesense-mcp-server, then add the server to your MCP client’s configuration file (e.g., ~/.cursor/mcp.json or Claude Desktop config). Set the required environment variables (TYPESENSE_HOST, TYPESENSE_PORT, TYPESENSE_PROTOCOL, TYPESENSE_API_KEY). Run the server with the default STDIO transport or switch to HTTP transports for web clients.

Key features of Typesense MCP Server

Checks Typesense server health and lists all collections
Creates, describes, exports, deletes, and truncates collections
Creates, upserts, deletes, and batch-indexes documents
Imports documents from CSV data into a collection
Performs keyword search and vector similarity search on collections
Supports STDIO, legacy SSE, and Streamable HTTP transports

Use cases of Typesense MCP Server

Build an AI assistant that can search a Typesense-powered product catalog
Manage collection schemas and bulk-import documents from a conversational interface
Enable semantic vector search queries on text or embeddings stored in Typesense
Automate index maintenance tasks (e.g., truncate, export, delete) via MCP clients

FAQ from Typesense MCP Server

How do I install the Typesense MCP Server?

Install uv (e.g., via Homebrew on macOS), clone the GitHub repository, then configure your MCP client to launch the server with uv run main.py and the required environment variables.

What environment variables are required?

You must set TYPESENSE_HOST, TYPESENSE_PORT, TYPESENSE_PROTOCOL, and TYPESENSE_API_KEY. Optional variables include MCP_TRANSPORT, MCP_HOST, MCP_PORT, MCP_STATELESS_HTTP, and MCP_CORS_ORIGINS.

What MCP transports does this server support?

It supports STDIO (default), legacy SSE, and the newer Streamable HTTP transport. STDIO works with desktop clients; Streamable HTTP is recommended for web-based clients.

Can I use this server with a web client?

Yes. Use the Streamable HTTP transport (set MCP_TRANSPORT=streamable-http) and enable CORS with MCP_CORS_ORIGINS. Stateless mode (MCP_STATELESS_HTTP=true) is required for clients that don’t maintain persistent MCP sessions.

Does this server require a running Typesense instance?

Yes. The server communicates with an existing Typesense server. You must provide its host, port, protocol, and API key via environment variables.

Comments

More Other MCP servers

Blender

ahujasid

Open-source MCP to use Blender with any LLM

23.6K

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K