Opensearch

🚀 MCP-Server-OpenSearch: An OpenSearch MCP Server

MCP-Server-OpenSearch enables seamless integration between LLM applications and OpenSearch, a distributed search and analysis engine, through the Model Context Protocol (MCP).

Model Context Protocol (MCP) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. Whether you're building an AI-driven IDE, enhancing a chat interface, or creating custom AI workflows, MCP provides a standard way to connect LLMs with their required context.

This repository demonstrates an example of creating an MCP server for OpenSearch, a distributed search and analysis engine.

🚧 Current Issue - Unable to Install OpenSearch Async Client

Open Search Async Client Documentation

pip install opensearch-py[async]
zsh: no matches found: opensearch-py[async]

✨ Overview

A basic Model Context Protocol server for storing and retrieving memories in the OpenSearch engine. It serves as a semantic memory layer on top of the OpenSearch database.

🧰 Components

Tools

1. search-openSearch
   • Store a memory in the OpenSearch database.
   • Input:
     • query (json): A prepared JSON query message.
   • Return: A confirmation message.

📦 Installation

Install via Smithery

To automatically install mcp-server-opensearch for Claude Desktop via Smithery:

npx -y @smithery/cli install @ibrooksSDX/mcp-server-opensearch --client claude

Using uv (Recommended)

When using uv, you can directly run mcp-server-opensearch without any specific installation.

uv run mcp-server-opensearch \
  --opensearch-url "http://localhost:9200" \
  --index-name "my_index" \

Or

uv run fastmcp run demo.py:main

💻 Testing - Local OpenSearch Client

uv run python src/mcp-server-opensearch/test_opensearch.py

💻 Testing - Connection between MCP Server and OpenSearch Client

cd src/mcp-server-opensearch
uv run fastmcp dev demo.py

🤝 Usage with Claude Desktop

To use this server with the Claude Desktop application, add the following configuration to the "mcpServers" section of claude_desktop_config.json:

{
  "opensearch": {
    "command": "uvx",
    "args": [
      "mcp-server-opensearch",
      "--opensearch-url",
      "http://localhost:9200",
      "--opensearch-api-key",
      "your_api_key",
      "--index-name",
      "your_index_name"
    ]
  }
}

Then run the following command for installation:

uv run fastmcp install

⚙️ Environment Variables

Here are the environment variables you can use and their functions:

• OPENSEARCH_URL: The URL address of the OpenSearch server.
• OPENSEARCH_API_KEY: The OpenSearch API key for authentication.
• OPENSEARCH_INDEX_NAME: The name of the index where data will be stored.

Make sure to set these environment variables before running the server.