zilliztech/mcp-server-milvus

Overview

This is an open-source Python server that implements the Model Context Protocol (MCP) for Milvus, a vector database. It allows AI agents and LLM applications to interact with Milvus collections, perform vector searches, and manage data through a standardized MCP interface.

Best for

Best for
Developers building AI agents or LLM applications that need to query and manage vector data in Milvus.

Use cases

• Connect an AI agent to a Milvus vector database for semantic search
• Manage Milvus collections and insert/query vectors from an LLM tool
• Build retrieval-augmented generation (RAG) pipelines using MCP-compatible clients

How to use

Tools exposed

• milvus_text_search
• milvus_vector_search
• milvus_hybrid_search

Tested with

Claude Desktop, Cursor

Example client config

[object Object]

Notes

This is an open-source Python server that implements the Model Context Protocol (MCP) for Milvus, a vector database. It allows AI agents and LLM applications to interact with Milvus collections, perform vector searches, and manage data through a standardized MCP interface.

232 stars on GitHub. Last updated 2026-05-08. Licensed Apache-2.0.

Use cases

• Connect an AI agent to a Milvus vector database for semantic search
• Manage Milvus collections and insert/query vectors from an LLM tool
• Build retrieval-augmented generation (RAG) pipelines using MCP-compatible clients

Pros

• Standardizes Milvus access for any MCP-compatible AI agent or framework
• Lightweight Python server with clear protocol boundaries
• Active open-source project with 232 GitHub stars and community support

Cons

• Requires a running Milvus instance (local or cloud) to function
• Limited to Milvus-specific operations, not a general-purpose database tool
• Dependency on the evolving MCP specification and client compatibility

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.