Typesense MCP Server

Typesense MCP Server is an MCP server that connects AI assistants to the Typesense search engine for administrative tasks, document management, and search querying. Built using the Model Context Protocol Python SDK, it enables software developers, data engineers, and AI agents to manage search indices and retrieve information without switching to external dashboards or writing custom API scripts. Users can query their Typesense instances using standard keyword search or vector similarity search directly within supported client conversations. In addition to retrieval, the server supports index lifecycle management, allowing assistants to check cluster health, inspect schemas, create collections, truncate indices, and delete collections. Data operations include creating, upserting, batch indexing, deleting individual documents, and importing structured records from CSV payloads. It supports STDIO, SSE, and Streamable HTTP transports, accommodating local developer environments as well as remote web client deployments.

Category: Databases & Data Stores

Tags: indexing, search, typesense, vector-search

Visit Typesense MCP Server

How to install and configure Typesense MCP Server

1. Ensure Python 3.11 or higher is installed, along with the uv package manager. 2. Clone the repository locally: shell git clone git@github.com:avarant/typesense-mcp-server.git ~/typesense-mcp-server 3. Add the server configuration to your MCP client config file (e.g., ~/.cursor/mcp.json or ~/Library/Application Support/Claude/claude_desktop_config.json): json { "mcpServers": { "typesense": { "command": "uv", "args": ["--directory", "~/typesense-mcp-server", "run", "mcp", "run", "main.py"], "env": { "TYPESENSE_HOST": "localhost", "TYPESENSE_PORT": "8108", "TYPESENSE_PROTOCOL": "http", "TYPESENSE_API_KEY": "your_api_key" } } } } 4. Restart your client to initialize the server tools.

What you can do with Typesense MCP Server

• Performing semantic and keyword queries across collections using the vector_search and search tools directly from conversational LLMs. - Inspecting Typesense cluster health and viewing index schemas to diagnose connectivity or schema validation issues. - Creating, truncating, or deleting collections when managing database schemas and prototyping new search indices. - Indexing single records, running bulk batch updates, or importing CSV data into target collections automatically. - Removing outdated or irrelevant documents by ID to maintain accurate index contents through agentic tool calling.

Key facts

• Open Source
• https://github.com/avarant/typesense-mcp-server
• Databases & Data Stores, Web Search & Research
• indexing, search, typesense, vector-search

Part of MCP Servers

How do I install Typesense MCP Server?

Install uv and ensure Python 3.11 or higher is available. Clone the repository to your machine using git clone. Then, register the server within your MCP client configuration file by pointing the command to uv and providing your Typesense host, port, protocol, and API key environment variables.

What can Typesense MCP Server do?

Typesense MCP Server exposes tools to monitor cluster health, list collections, create or delete schemas, and truncate indices. For content, it allows adding, upserting, deleting, and batch indexing documents, as well as importing records from CSV. It also executes keyword and vector similarity searches across specified collections.

Which MCP clients work with Typesense MCP Server?

The server works with any MCP-compliant client. It supports STDIO for desktop applications like Claude Desktop, Cursor, Windsurf, Zed, and VS Code. It also supports SSE and Streamable HTTP transports for browser-based clients and remote architectures.

Is Typesense MCP Server open source?

Yes, Typesense MCP Server is open source software. The repository is publicly hosted on GitHub under the avarant organization, allowing developers to inspect the source code, contribute enhancements, or customize tools.