Qdrant MCP Server

Implement semantic memory layer on top of the Qdrant vector search engine

Local serverstdioPython

What is the Qdrant MCP server?

Qdrant mcp server connects Qdrant to AI assistants that speak the Model Context Protocol. Implement semantic memory layer on top of the Qdrant vector search engine.

What Qdrant does

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

An official Model Context Protocol server for keeping and retrieving memories in the Qdrant vector search engine. It acts as a semantic memory layer on top of the Qdrant database.

Tools it exposes

Once connected, the assistant can call these 3 tools directly:

  • Input — - information (string): Information to store
  • Returns — Confirmation message
  • Tools — 1. qdrant-store - Store some information in the Qdrant database - Input: - information (string): Information to store - metadata (JSON): Optional metadata to

Installing the qdrant mcp server

The server is distributed via npm as @smithery/cli, so most clients can run it without a manual build step. Add it to your MCP client's configuration and restart the client to pick it up — the copy-paste configs for Claude Desktop, Claude Code and Cursor are on this page.

Configuration

Before the server will start you need to supply 6 environment variables: QDRANT_URL, QDRANT_API_KEY, COLLECTION_NAME, EMBEDDING_MODEL, QDRANT_LOCAL_PATH, FASTMCP_SERVER_HOST. Keep credentials in your client's env block or a secrets manager rather than committing them.

Where it fits

Database servers turn schema archaeology and ad-hoc reporting into conversation, which is why they tend to be the second or third server people install. Qdrant sits in that group, and the shape of its toolset — Input, Returns, Tools — tells you what it is really for. Worth comparing against the other databases servers in this directory before you commit to one, since several overlap in scope but differ sharply in setup cost and permissions.

Practical notes

  • This server runs locally, so it operates with whatever access your machine and its credentials already have. Scope that deliberately rather than by default.
  • It will not start until its required credentials are present, so set those before wondering why the tools never appear.
  • Written in Python.
  • MCP clients ask for confirmation before each tool call by default. Keep that on while you learn what the qdrant mcp server actually does with your data.
  • Every entry in this directory is reviewed by hand before it goes live, and details are checked against the project's own documentation.

Available tools

ToolWhat it does
Input- information (string): Information to store
ReturnsConfirmation message
Tools1. qdrant-store - Store some information in the Qdrant database - Input: - information (string): Information to store - metadata (JSON): Optional metadata to store - collection_name (string): Name of the collection to st

How to install the Qdrant MCP server

{
  "mcpServers": {
    "server-qdrant": {
      "command": "npx",
      "args": ["-y", "@smithery/cli"],
      "env": {
        "QDRANT_URL": "your-value",
        "QDRANT_API_KEY": "your-value",
        "COLLECTION_NAME": "your-value",
        "EMBEDDING_MODEL": "your-value",
        "QDRANT_LOCAL_PATH": "your-value",
        "FASTMCP_SERVER_HOST": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

VariableDescriptionRequired
QDRANT_URLEndpoint or connection string the server talks to.Yes
QDRANT_API_KEYCredential the server authenticates with.Yes
COLLECTION_NAMEConfiguration value read at startup.Optional
EMBEDDING_MODELConfiguration value read at startup.Optional
QDRANT_LOCAL_PATHFilesystem location the server is allowed to use.Optional
FASTMCP_SERVER_HOSTEndpoint or connection string the server talks to.Optional

Example prompts to try

  • Use Qdrant to Input.
  • Use Qdrant to Returns.
  • Use Qdrant to Tools.

Frequently asked questions

It connects Qdrant to MCP-compatible AI assistants such as Claude and Cursor, exposing 3 tools (Input, Returns, Tools) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Qdrant directly.