Milvus MCP Server Integration

What does “Milvus” MCP Server do?

The Milvus MCP (Model Context Protocol) Server connects AI assistants and LLM-powered applications with the Milvus vector database. This enables seamless interaction between language models and large-scale vector data, providing a standardized way to access, query, and manage Milvus from within AI workflows. Using the Milvus MCP Server, developers can integrate Milvus-based search, retrieval, and data management capabilities directly into their AI agents, IDEs, or chat interfaces. The server supports multiple communication modes (stdio and Server-Sent Events), allowing it to fit diverse deployment scenarios and development environments. By bridging LLMs and Milvus, it greatly enhances the ability of AI systems to perform context-aware operations on high-dimensional data, unlocking richer and more intelligent LLM-powered experiences.

List of Prompts

No information about prompt templates is provided in the repository.

List of Resources

No explicit list of Model Context Protocol “resources” is described in the available documentation or code.

List of Tools

No explicit tool list or function names are documented in the available documentation or code files, including server.py.

Use Cases of this MCP Server

• Vector Search Integration: Enables developers to use LLMs to query and retrieve relevant documents or data points from Milvus, enhancing contextual search in AI applications.
• Embedding Management: Allows LLMs and agents to store and manage vector embeddings within Milvus, supporting advanced semantic search workflows.
• Chatbot Contextual Memory: Facilitates chatbots or AI assistants in maintaining long-term memory by storing conversational data as vectors in Milvus for later retrieval.
• Data Analysis and Recommendation: Powers AI-driven recommendation systems by allowing LLMs to perform similarity searches over large datasets stored in Milvus.
• Real-time Data Access: Supports AI agents that require real-time access to high-dimensional data for analytics, pattern recognition, or anomaly detection.

How to set it up

Windsurf

1. Ensure you have Python 3.10+ and a running Milvus instance.
2. Clone the repository:
   git clone https://github.com/zilliztech/mcp-server-milvus.git
3. Run the server:
   uv run src/mcp_server_milvus/server.py --milvus-uri http://localhost:19530
4. Add the MCP server to your Windsurf configuration:

{
  "mcpServers": {
    "milvus-mcp": {
      "command": "uv",
      "args": ["run", "src/mcp_server_milvus/server.py", "--milvus-uri", "http://localhost:19530"]
    }
  }
}

5. Save and restart Windsurf. Verify connection in the interface.

Securing API keys:
If the server requires sensitive info, use environment variables:

{
  "env": {
    "MILVUS_URI": "http://localhost:19530"
  },
  "inputs": {}
}

Claude

1. Install prerequisites: Python 3.10+, Milvus, and uv.
2. Clone and start the server as described above.
3. In Claude’s settings, add the MCP server with:

{
  "mcpServers": {
    "milvus-mcp": {
      "command": "uv",
      "args": ["run", "src/mcp_server_milvus/server.py", "--milvus-uri", "http://localhost:19530"]
    }
  }
}

4. Save and restart Claude. Confirm Milvus MCP appears in available tools.

Secure credentials via environment variables as above.

Cursor

1. Install Python 3.10+ and Milvus, plus uv.
2. Clone the repo and run:
   uv run src/mcp_server_milvus/server.py --milvus-uri http://localhost:19530
3. In Cursor’s configuration, add:

{
  "mcpServers": {
    "milvus-mcp": {
      "command": "uv",
      "args": ["run", "src/mcp_server_milvus/server.py", "--milvus-uri", "http://localhost:19530"]
    }
  }
}

4. Restart Cursor and verify setup.

Securing API keys:
Use environment variables as shown above.

Cline

1. Prerequisites: Python 3.10+, Milvus, and uv.
2. Clone the repository and start the server.
3. Edit Cline’s config to add:

{
  "mcpServers": {
    "milvus-mcp": {
      "command": "uv",
      "args": ["run", "src/mcp_server_milvus/server.py", "--milvus-uri", "http://localhost:19530"]
    }
  }
}

4. Save changes and restart Cline.

Environment variables:

{
  "env": {
    "MILVUS_URI": "http://localhost:19530"
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "milvus-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “milvus-mcp” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Roots support: Not mentioned
Sampling support: Not mentioned

Our opinion

The Milvus MCP Server is a practical and focused bridge for connecting LLMs with Milvus, with clear setup guides for popular dev tools. However, its documentation lacks detail on MCP resources, prompts, and actionable tool APIs, which limits out-of-the-box discoverability. Still, it’s a solid foundation for vector-based AI integrations.

MCP Score

Overall: 4/10
The server is useful for its niche but would benefit greatly from more explicit documentation on resources, prompt templates, and tool APIs for maximal interoperability and ease of use.