Meilisearch

Meilisearch MCP Server provides a Model Context Protocol interface to interact with Meilisearch. It enables you to manage indexes, documents, searches, and settings through a consistent, programmable API ideal for AI assistants and automation.

How to use

You connect to the Meilisearch MCP Server from an MCP client to perform common data operations and queries. Use it to create and modify indexes, add or update documents, run searches with filters, adjust index settings, and monitor tasks. You can also enable experimental vector search features and perform health and status checks to ensure your Meilisearch instance remains healthy.

How to install

Prerequisites: Node.js and npm must be available on your machine. You may also choose to run the server in a container using Docker Compose if you prefer containerization.

# Clone the MCP server repository
git clone https://github.com/devlimelabs/meilisearch-ts-mcp.git
cd meilisearch-ts-mcp

# Install dependencies
npm install

# Optional: copy example environment and customize
cp .env.example .env

Configuration and usage notes

You can run the MCP server locally and connect to it from MCP clients such as Cursor or Claude Desktop. Build and host details are provided for local development and client integrations.

# Build the project for local development
npm run build

# Start the MCP server in development mode
npm start

Claude Desktop integration

To integrate Claude Desktop with the Meilisearch MCP Server, you can use an automated setup script or configure the client manually.

# Build the project
npm run build

# Run the Claude Desktop setup script
node scripts/claude-desktop-setup.js

Cursor integration

You can connect Cursor to the MCP server by starting the MCP server locally and then configuring Cursor to run the server process with the appropriate environment variables.

# Start the MCP server (example path; adjust to your environment)
# Ensure the server is running at the path you provide

Available tools

create-index

Create a new index with the given name and settings.

get-index

Retrieve information about a specific index.

list-indexes

List all indexes available in the Meilisearch instance.

update-index

Update the configuration of an existing index.

delete-index

Delete a specified index.

add-documents

Add documents to a specified index.

get-document

Retrieve a single document by ID from an index.

get-documents

Retrieve multiple documents by IDs from an index.

update-documents

Update documents within an index.

delete-document

Delete a single document by ID from an index.

delete-documents

Delete multiple documents from an index.

delete-all-documents

Delete all documents from an index.

search

Execute a search against an index with optional parameters and filters.

multi-search

Perform multiple searches in a single request.

get-settings

Retrieve current settings for an index.

update-settings

Update index settings such as synonyms, stop words, and ranking rules.

reset-settings

Reset index settings to their default values.

enable-vector-search

Enable vector search capabilities for an index.

get-experimental-features

Query the status of experimental features.

update-embedders

Configure embedders for vector representations.

get-embedders

Retrieve the current embedders configuration.

reset-embedders

Reset embedders configuration to defaults.

vector-search

Perform a vector-based search on an index.