Elasticsearch MCP
About
Elastic's official MCP server lets agents query Elasticsearch in natural language — list indices, view mappings, run DSL and ES|QL queries, do full-text and vector search, bringing search-engine data straight into AI conversations.
Our Verdict
Elasticsearch MCP brings database capabilities to the MCP ecosystem. Its GitHub popularity reflects community trust, and the standardized protocol ensures broad compatibility across AI clients.
Pricing
Use Cases
- Integrating database capabilities into AI agent workflows
- Building custom AI-powered tools with Elasticsearch MCP
- Enhancing AI agent functionality with standardized MCP access
Pros
- Elasticsearch MCP provides standardized MCP protocol access
- Open source with 700 GitHub stars
- Active development and community support
Cons
- Requires MCP-compatible client (Claude, Cursor, etc.)
- Setup complexity varies by client
- Some features may depend on client capabilities
Installation
npx -y @elastic/mcp-server-elasticsearchFrequently Asked Questions
What can Elasticsearch MCP do?
It lets AI explore your Elasticsearch cluster in natural language: list indices, view field mappings, run Query DSL or the more concise ES|QL, and do full-text and vector-similarity search. Great for troubleshooting and analytics where you "ask your logs or business data" in chat.
Does it support Elastic Cloud and self-hosted clusters? How does auth work?
Both are supported. Pass the cluster URL (ES_URL) and credentials via environment variables: use an API key for Elastic Cloud, or an API key or username/password for self-hosted. Prefer a read-only API key to keep the agent from altering indices.
Is Elasticsearch MCP free?
The MCP server is officially open-sourced by Elastic and free (Apache-2.0); it queries your own cluster. Local or self-hosted Elasticsearch costs nothing, while Elastic Cloud is billed per its subscription.