Model Context Protocol - Talk to Meilisearch with Claude desktop

This guide will walk you through setting up and using Meilisearch through natural language interactions with Claude AI via Model Context Protocol (MCP).

To follow this guide, you'll need:

• Claude Desktop (free)
• A Meilisearch Cloud project (14 days free-trial)
• Python ≥ 3.9
• From the Meilisearch Cloud dashboard, your Meilisearch host & api key

Download and install Claude Desktop.

You can install the Meilisearch MCP server using uv or pip:

# Using uv (recommended)
uv pip install meilisearch-mcp

# Using pip
pip install meilisearch-mcp

Open Claude Desktop, click on the Claude menu in the top bar, and select "Settings". In the Settings window, click on "Developer" in the left sidebar, then click "Edit Config". This will open your claude_desktop_config.json file.

Add the Meilisearch MCP server to your configuration:

{
  "mcpServers": {
    "meilisearch": {
      "command": "uvx",
      "args": ["-n", "meilisearch-mcp"]
    }
  }

Save the file and restart Claude.

Once Claude Desktop is set up with the Meilisearch MCP server, you can connect to your Meilisearch instance by asking Claude to update the connection settings.

Open Claude Desktop and start a new conversation.

Next, connect to your Meilisearch instance by asking Claude to update the connection settings, replacing MEILISEARCH_URL with your project URL and API_KEY with your project's API key:

Please connect to my Meilisearch instance at MEILISEARCH_URL using the API key API_KEY

Claude will use the MCP server's update-connection-settings tool to establish a connection to your Meilisearch instance.

Finally, verify the connection by asking:

Can you check the connection to my Meilisearch instance and tell me what version it's running?

Claude will use the get-version and health-check tools to verify the connection and provide information about your instance.

Now you have configured the MCP to work with Meilisearch, you can use it to manage your indexes.

First, verify what indexes you have in your project:

What indexes do I have in my Meilisearch instance?

Next, ask Claude to create an index optimized for e-commerce:

Create a new index called "products" for our e-commerce site with the primary key "product_id"

Finally, check the index has been created successfully and is completely empty:

How many documents are in my "products" index and what's its size?

Ask Calude to add a couple of test documents to your "products" index:

Add these products to my "products" index:
[
  {"product_id": 1, "name": "Ergonomic Chair", "description": "Comfortable office chair", "price": 299.99, "category": "Furniture"},
  {"product_id": 2, "name": "Standing Desk", "description": "Adjustable height desk", "price": 499.99, "category": "Furniture"}
]

Since you are only using "products" for testing, you can also ask Claude to automatically populate it with placeholder data:

Add 10 documents in the index "products" with a name, category, price, and description of your choice

To verify data insertion worked as expected, retrieve the first few documents in your index:

Show me the first 5 products in my "products" index

Before performing your first search, set a few index settings to ensure relevant results.

Ask Claude to prioritize exact word matches over multiple partial matches:

Update the ranking rules for the "products" index to prioritize word matches and handle typos, but make exact matches more important than proximity

It's also a good practice to limit searchable attributes only to highly-relevant fields, and only return attributes you are going to display in your search interface:

Configure my "products" index to make the "name" and "description" fields searchable, but only "name", "price", and "category" should be displayed in results

Perform your first search with the following prompt:

Search the "products" index for "desk" and return the top 3 results

You can also request your search uses other Meilisearch features such as filters and sorting:

Search the "products" index for "chair" where the price is less than 200 and the category is "Furniture". Sort results by price in ascending order.

If you encounter issues with the Meilisearch MCP integration, try these steps

What are the current Meilisearch connection settings?

Run a health check on my Meilisearch instance

Open the logs file in your text editor or log viewer:

• On macOS: ~/Library/Logs/Claude/mcp*.log
• On Windows: %APPDATA%\Claude\logs\mcp*.log

Open your terminal and query the MCP Inspector with npx:

npx @modelcontextprotocol/inspector uvx -n meilisearch-mcp

The Meilisearch MCP integration with Claude can transform multiple API calls and configuration tasks into conversational requests. This can help you focus more on building your application and less on implementation details.

For more information about advanced configurations and capabilities, refer to the Meilisearch documentation and the Meilisearch MCP server repository.