meilisearch-mcp

Overview

The Meilisearch MCP Server is a Python-based implementation that allows MCP-compatible clients to manage search indices, perform searches, and handle data through conversation. It provides comprehensive functionality including index management, document operations, advanced search capabilities with filtering and sorting, settings configuration, and system monitoring. The server uses stdio transport and offers both Python and TypeScript implementations.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• get-connection-settings

  View current Meilisearch connection URL and API key status

  Try:Show me the current connection settingsTry:What's the current Meilisearch URL and API key?
• update-connection-settings

  Update URL and/or API key to connect to a different instance

  Try:Update connection to use the staging Meilisearch serverTry:Change the API key for production access
• create-index

  Create a new index with optional primary key

  Try:Create a new index called 'products' with 'id' as the primary keyTry:Make a new index for blog posts
• list-indexes

  List all available indexes

  Try:Show me all available indexesTry:List the indexes with their stats
• delete-index

  Delete an existing index and all its documents

  Try:Delete the old_products indexTry:Remove the temporary_test index
• get-documents

  Retrieve documents from an index with pagination

  Try:Get documents from the products indexTry:Show me the first 10 documents in the blog_posts index
• add-documents

  Add or update documents in an index

  Try:Add these product documents to the products indexTry:Update the user documents with new information
• search

  Flexible search across single or multiple indices with filtering and sorting

  Try:Search for 'laptop' in the products indexTry:Find articles about machine learning published after 2023
• get-settings

  View current settings for an index

  Try:Show me the current settings for the products indexTry:What are the ranking rules for the blog index?
• update-settings

  Update index settings (ranking, faceting, etc.)

  Try:Update the settings to enable typo toleranceTry:Change the ranking rules for the search index
• get-tasks

  List tasks with optional filters

  Try:Show me the latest indexing tasksTry:List all failed tasks
• health-check

  Basic health check of the Meilisearch instance

  Try:Is the Meilisearch instance healthy?Try:Check the status of the search server

Comparable tools

Installation

pip install meilisearch-mcp

uvx -n meilisearch-mcp

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

docker pull getmeili/meilisearch-mcp:latest

docker run -it \
  -e MEILI_HTTP_ADDR=http://your-meilisearch:7700 \
  -e MEILI_MASTER_KEY=your-master-key \
  getmeili/meilisearch-mcp:latest

FAQ

What MCP clients does this server support?

The Meilisearch MCP Server works with any MCP-compatible client, including Claude, OpenAI agents, and other LLMs.

How do I connect to a different Meilisearch instance?

You can update the connection settings dynamically through the MCP tools or by setting the MEILI_HTTP_ADDR and MEILI_MASTER_KEY environment variables.

Compare meilisearch-mcp with