- Home
- MCP servers
- Meilisearch
Meilisearch
- go
0
GitHub Stars
go
Language
7 months ago
First Indexed
3 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"cnosuke-mcp-meilisearch": {
"command": "./bin/mcp-meilisearch",
"args": [
"server",
"--no-logs",
"--log",
"path_to_log_file"
],
"env": {
"MEILISEARCH_HOST": "http://localhost:7700",
"MEILISEARCH_API_KEY": "api_key"
}
}
}
}You deploy this MCP server to expose Meilisearch search capabilities to large language models. It wraps Meilisearch in an MCP-compatible server, letting you perform searches, manage indexes, and handle documents through a standardized protocol that your MCP client can consume.
How to use
You run the MCP server locally and connect your MCP client to it. The server speaks through stdio, so you launch it in the background and direct logs to a file to keep the protocol messages clean. From your MCP client, you can perform common actions such as health checks, indexing, and searching. Use the provided tool endpoints to check server status, list and create indexes, add documents, and execute search queries against a specific index.
How to install
Prerequisites: install Go 1.24 or higher and ensure a running Meilisearch instance.
# Clone the repository
git clone https://github.com/cnosuke/mcp-meilisearch.git
cd mcp-meilisearch
# Install dependencies
make deps
# Build the MCP server binary
make bin/mcp-meilisearch
Configuration and startup
Configure the connection to Meilisearch in a YAML file. You can also provide these settings via environment variables when starting the server.
meilisearch:
host: http://localhost:7700 # Address of Meilisearch server
api_key: '' # Set API key if needed
Starting the server
Start the MCP server using the built binary and supply the configuration file path.
./bin/mcp-meilisearch server --config config.yml
Using with Claude Desktop
If you use Claude Desktop, register an MCP server entry and direct logs away from stdio by using log redirection flags. The example shows how to pass the Meilisearch connection details via environment variables.
{
"mcpServers": {
"sqlite": {
"command": "./bin/mcp-meilisearch",
"args": ["server", "--no-logs", "--log", "path_to_log_file"],
"env": {
"MEILISEARCH_HOST": "http://localhost:7700",
"MEILISEARCH_API_KEY": "api_key"
}
}
}
}
Notes on logging and MCP stdio
Because MCP uses stdio for communication, redirect logs away from stdio so messages remain clean. Use the --no-logs flag and specify a log file with --log.
Security considerations
Protect your Meilisearch API key and restrict access to the MCP server to trusted clients. If you expose the HTTP endpoint, use a secure channel and consider API key restrictions on the Meilisearch side.
Migration and troubleshooting
If you encounter startup issues, verify that Meilisearch is reachable at the host URL and that the API key, if used, is correct. Check the log file you provided with --log for any error messages during server initialization or runtime.
Available tools
health_check
Check the status of the Meilisearch server.
list_indexes
Retrieve a list of all indexes.
create_index
Create a new index with a unique identifier and optional primary key.
get_documents
Retrieve documents from an index with optional pagination and field selection.
add_documents
Add documents to an index and receive an addition task status.
search
Search for documents in an index with optional limit, offset, filter, and sort.