Elasticsearch MCP server: setup, auth, production

What is the Elasticsearch MCP server?

A Elasticsearch MCP server implements the Model Context Protocol and exposes Elasticsearch's capabilities as typed tools an agent can discover and call in the middle of a task. Your search and log analytics cluster.

Warehouse access turns an agent into an analyst that writes and explains its own queries. The governance that matters here is cost (bytes scanned / warehouse time) and row-level access to sensitive tables — both are solvable, neither is optional.

A real Elasticsearch workflow

Mid-incident, the agent queries the logs directly — narrowing to the error signature, the affected shard, the time window — and explains what it found, instead of you hand-building a Kibana query under pressure.

What an agent can do with Elasticsearch

• Let an agent query logs while debugging an incident
• Draft and explain a search query
• Investigate index health and mappings

Read-only is the right default: let the agent observe and reason first, then grant specific write actions deliberately — each behind audit logging and, for anything high-impact, human approval.

Setting up the Elasticsearch MCP server

Issue an API key with a read-only role limited to the relevant indices. Elasticsearch queries can be expensive; restrict the key’s indices and consider a search-slow-log to catch an agent’s heavy query.

Auth: An API key with read-only role privileges. Credentials belong in the environment or a secret store — never in the repo or the agent's sandbox. Wiring a server into Claude Code or Cursor is the same for every tool; the step-by-step lives in the MCP connection guide.

The risk to watch with Elasticsearch

A cluster-wide, unbounded query from an agent can degrade search for everyone. Index-scoped read-only keys and query cost awareness are the controls.

More broadly, a production-grade Elasticsearch integration needs read-only roles and guarding against expensive cluster-wide queries. The MCP security checklist applies to every integration — we build servers with these controls baked in.