Elastic Cloud
Give the Sazabi agent tools to search Elasticsearch indices and query Agent Builder tools from your Kibana-hosted MCP endpoint.
All Elastic Cloud integration capabilities
About
This is an MCP connector that gives the Sazabi agent tools from Elastic Cloud's built-in MCP server, hosted at /api/agent_builder/mcp on your Kibana instance. Sazabi connects using an Elastic API key, which it stores securely and sends as the Authorization header to your Kibana-based MCP endpoint. Because the server URL comes from your own Kibana deployment, you supply the Kibana base URL and Sazabi builds the full endpoint.
What the agent can do:
- Search Elasticsearch indices using full Query DSL
- List available indices and retrieve field mappings
- Fetch documents by ID
- Call custom Agent Builder tools you have defined in Kibana
- Analyze logs, metrics, and traces stored in Elasticsearch
Prerequisites
- An Elastic Cloud account on version 9.2+ or a Serverless project
- Agent Builder enabled in your Kibana AI Assistant configuration
- Your Kibana base URL. Go to Elastic Cloud deployments, open the deployment you want to connect, open Kibana, then copy the Kibana base URL from the browser.
- An Elastic API key with the following privileges:
- Kibana application:
feature_agentBuilder.readandfeature_actions.readonkibana-.kibana - Cluster:
monitor_inference - Index:
readandview_index_metadataon the indices you want agents to access
- Kibana application:
Set up in the dashboard
Configure connectors in the dashboard under Settings > MCP connectors.
Open the connector
Find Elastic Cloud in Settings > MCP connectors.
Add your key
Sazabi needs two values:
-
Kibana URL — Go to Elastic Cloud deployments, open the deployment you want to connect, then open Kibana and copy the Kibana base URL from the browser. If your Kibana URL contains
/s/<space>/, note the space name for the optional Kibana space field. -
Elastic API key — In Kibana, go to Stack Management > API keys and create an API key with the Agent Builder MCP server privileges listed in the prerequisites. Copy the generated key.
Sazabi stores the key securely and sends it as the Authorization header to Elastic Cloud's hosted MCP endpoint, built from your Kibana base URL as <kibana-url>/api/agent_builder/mcp.
Set up with the CLI
Elastic Cloud authenticates with an Elastic API key. With the Sazabi CLI installed and authenticated (see sazabi mcp-connectors reference), connect it in one command, passing the credential as a header. Elastic Cloud has no default server, so --server-url names your deployment's MCP endpoint:
sazabi mcp-connectors connect elastic-cloud --server-url <your-server-url> --header "Authorization=ApiKey <api-key>"The CLI connects to the MCP server at once and prints the connection with the tools it discovered; a credential the server rejects is reported, and the connection is left needing attention. Use NAME=@<file> to read a value from a file, or --header-file <file.json> for a JSON object of header names and values. To rotate the stored credentials later, use sazabi mcp-connectors update-credentials <connection-id>; find the connection ID with sazabi mcp-connectors list.
Verify
Open the connector's connection page under Settings > MCP connectors and confirm Sazabi lists the tools it discovered. Then, in a thread, ask the agent to search an index and confirm it returns results. If no tools appear, confirm Agent Builder is enabled and the API key has the required privileges.
Troubleshooting
No tools discovered — Confirm Agent Builder is enabled in your Kibana AI Assistant configuration, the API key has the required privileges, and the Kibana URL is correct.
Connection shows needs attention — The API key may have expired or been revoked; create a new key and reconnect from Settings > MCP connectors.