Typesense MCP Server

What is Typesense MCP Server?

The Typesense MCP Server is a Model Context Protocol (MCP) server that interfaces with Typesense, an open-source search engine. It provides tools to manage collections, documents, and perform search operations, intended for developers integrating Typesense with MCP-compatible clients.

How to use Typesense MCP Server?

Install uv, clone the repository, and configure the server in your MCP client (e.g., Cursor’s mcp.json) by setting the environment variables TYPESENSE_HOST, TYPESENSE_PORT, TYPESENSE_PROTOCOL, and TYPESENSE_API_KEY. Then run the server via the uv command.

Key features of Typesense MCP Server

• Check Typesense server health.
• List, describe, create, delete, and truncate collections.
• Export all documents from a collection.
• Create, upsert, and delete single documents.
• Index multiple documents in batch.
• Import documents from CSV data.
• Perform keyword and vector similarity search.

Use cases of Typesense MCP Server

• Manage Typesense collections and documents from an MCP client.
• Import CSV data into a Typesense collection effortlessly.
• Execute keyword or vector search queries programmatically.
• Automate indexing of multiple documents in a single batch.

FAQ from Typesense MCP Server

What is the Typesense MCP Server?

It is an MCP server that provides tools to interact with Typesense, allowing you to manage collections, documents, and perform searches through the Model Context Protocol.

What dependencies are required?

You need Python, the uv package manager, and a running Typesense server instance. The server uses environment variables to connect to Typesense.

How do I configure the server?

Set the environment variables TYPESENSE_HOST, TYPESENSE_PORT, TYPESENSE_PROTOCOL, and TYPESENSE_API_KEY in your MCP client configuration.

What tools are available?

Tools are grouped into server management (health check, list collections), collection management (describe, export, create, delete, truncate), document operations (create, upsert, batch index, delete, CSV import), and search (keyword search, vector search).

What transport does it use?

—