get-connection-settingsLABSCO SUMMARY
Meilisearch's own control surface, wide enough to build the index and then query it.
Most search integrations only let an agent read. This one covers creating the index, loading documents into it, tuning relevancy and issuing API keys, so the whole loop happens in one session. The key you hand it decides how much of that loop the agent can actually reach — give it a scoped key rather than the master key when the job is only search.
What it is
Meilisearch's own server for its search engine. It connects to an instance you point it at and hands over the whole thing — indexes, documents, settings, API keys and the task queue — not just a search box.
What you get
- Search across a single index or several at once, with filtering, sorting and facets — search
- Indexes created, listed and deleted, with per-index metrics — create-index, list-indexes, delete-index, get-index-metrics
- Documents added or updated, and read back with pagination — add-documents, get-documents
- Index settings read and changed: ranking rules, faceting, searchable attributes, typo tolerance — get-settings, update-settings
- API keys created, listed and deleted with specific permissions — create-key, get-keys, delete-key
- Indexing work tracked and queued tasks cancelled — get-task, get-tasks, cancel-tasks
- Version, health, database statistics and system information — get-version, health-check, get-health-status, get-stats, get-system-info
- The server repointed at a different Meilisearch instance mid-session — get-connection-settings, update-connection-settings
- Every data-returning tool ships structuredContent alongside its text, so a client can read the result as data rather than parse prose
Requirements
A running Meilisearch instance and Python 3.10 or newer; it starts through uvx, pip or the published Docker image. Two environment variables point it at the instance: MEILI_HTTP_ADDR for the URL and MEILI_MASTER_KEY for a key with the permissions you want the agent to have. Both can also be swapped mid-session — convenient in development, and the project's own advice is not to rely on that in production. MIT.
Setup effort
One command — uvx -n meilisearch-mcp
USE CASES
01
Standing up a new search index
Create the index with its primary key, load the documents, and read a page of them back to confirm what landed.
02
Answering a question across your own data
Search one index or several at once with filters and sorting, without writing a query against the REST API.
03
Fixing bad relevancy
Read the index's current ranking rules and searchable attributes, change them, then run the query that was wrong to see whether it moved.
04
Waiting on a big import
Watch the task queue while indexing runs, read one task in detail, and cancel what is still enqueued if the import was wrong.
05
Handing out scoped access
Create an API key limited to the actions and indexes it needs, check what already exists, and delete the ones that outlived their purpose.
06
Checking an instance is healthy
Health, version, database statistics and per-index metrics in the same session where the search that looked slow was run.
22 tools
Listed in project docs. Your client may show a different list.
update-connection-settingscreate-indexlist-indexesdelete-indexget-index-metricsget-documentsadd-documentssearchget-settingsupdate-settingsget-keyscreate-keydelete-keyget-taskget-taskscancel-taskshealth-checkget-health-statusget-versionget-statsget-system-infoBEFORE YOU INSTALL
What to know first
Tools you get
22 tools named in the project's documentation.
Requires
- uvx on PATH
- Python ≥ 3.10
- A running Meilisearch instance (the server connects to http://localhost:7700 unless MEILI_HTTP_ADDR says otherwise)
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
uvx -n meilisearch-mcpMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"meilisearch": {
"command": "uvx",
"args": [
"-n",
"meilisearch-mcp"
]
}
}
}As written, this connects to a Meilisearch instance at http://localhost:7700 with no key (start one with docker run -d -p 7700:7700 getmeili/meilisearch:v1.28). For another instance, add an "env" section to the meilisearch entry with MEILI_HTTP_ADDR set to its URL and MEILI_MASTER_KEY set to a key for it.
