Get started

Kibana for AI agents

Kibana is the visualization layer for Elasticsearch, giving engineering and security teams dashboards, search, and alerting over their log and event data. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.

Connect KibanaSee the 47 actions ↓
AnalyticsAPI keyMCP + RESTelastic.co
AUDIT LOG · KIBANAPOLICY: acme-support
09:41:07 · claude · u_8f2read
kibana.get_cases
Get Cases✓ allowed · 212ms
09:41:08 · claude · u_8f2read
kibana.get_status
Get Kibana Status✓ allowed · 164ms
09:41:09 · claude · u_8f2write
kibana.post_cases
Create Case✓ approved · approved by user
09:41:12 · claude · u_8f2destructive
kibana.delete_list
Delete List✕ blocked · policy: deny
EVERY KIBANA CALL, ON THE RECORD
01 · USE CASES

What agents do in Kibana.

01

Check open alerting rules

List the alerting rules currently configured for a data view before an on-call review.

02

Look up detection engine findings

Search detection engine rules and cases to summarize open security findings for a team standup.

03

Confirm before deleting a saved object

Require approval before removing a saved dashboard or connector, since deleting a saved object cannot be undone.

02 · ACTIONS

47 Kibana actions, graded by risk.

Every Kibana action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

34

Look things up. Allowed by default.

  • kibana.get_cases
    Get Cases
  • kibana.get_status
    Get Kibana Status
  • kibana.find_alerts
    Find Kibana Alerts
  • kibana.get_metrics
    Get Node Metrics
  • kibana.get_connectors
    Get All Connectors
  • kibana.get_data_views
    Get Data Views
  • kibana.get_alert_types
    Get Rule Types
  • kibana.get_action_types
    Get Action Types
  • kibana.get_saved_objects
    Get Saved Objects
  • kibana.get_alerting_rules
    Get Alerting Rules
  • kibana.get_reporting_jobs
    Get Reporting Jobs
  • kibana.get_fleet_server_host
    Get Fleet Server Host
  • kibana.get_fleet_epm_packages
    Get Fleet EPM Packages
  • kibana.get_fleet_server_hosts
    Get Fleet Server Hosts
  • kibana.get_endpoint_list_items
    Get Endpoint List Items
  • kibana.get_entity_store_status
    Get Entity Store Status
+ 18 MORE

write

6

Create and change things. Allow, or ask the user first.

  • kibana.post_cases
    Create Case
  • kibana.post_connectors
    Create Kibana Connector
  • kibana.post_dashboards
    Create Dashboard
  • kibana.post_data_views
    Create Data View
  • kibana.post_saved_objects
    Create or Update Saved Object
  • kibana.post_alerting_rules
    Create Alerting Rule

destructive

7

Delete, cancel or archive. Ask first, or deny outright.

  • kibana.delete_list
    Delete List
  • kibana.delete_connectors
    Delete Connector
  • kibana.delete_fleet_proxy
    Delete Fleet Proxy
  • kibana.delete_fleet_output
    Delete Fleet Output
  • kibana.delete_saved_objects
    Delete Saved Object
  • kibana.delete_alerting_rules
    Delete Alerting Rule
  • kibana.delete_osquery_saved_queries
    Delete Osquery Saved Query
03 · HOW IT WORKS

Kibana in three steps.

  1. 01Your users connect KibanaThey add their Kibana api key on Arc0 Connect, under your brand. It goes straight into the vault.
  2. 02You set the rulesReads run, writes like “create Case” can wait for the user, and “delete List” can be denied outright.
  3. 03Any agent can actYour agent calls Kibana through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
POLICY.TS
await arc0.policies.set('kibana', {
  read: 'allow',
  write: 'ask',        // post_cases
  destructive: 'deny',  // delete_list
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2
04 · AUTH AND DATA

How Kibana connects.

Users add their Kibana api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

The same Kibana connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →

AUTH
API key · Username and password
CREDENTIALS
Per-tenant encrypted vault
MODEL SEES
Results only, never credentials
AUDIT LOG
Every call, on every plan
05 · WORKS WITH

Use Kibana from any agent.

Claude
ChatGPT
Cursor
Codex
VS Code
OpenAI Agents SDK
Claude Agent SDK
Vercel AI SDK
Mastra
LangGraph
06 · RELATED

More analytics apps.

Gong58 actionsSerpApi59 actionsPeople Data Labs27 actionsPostHog507 actions21risk8 actionsAkta19 actionsAlpha Vantage56 actionsAmplitude54 actions
ALL ANALYTICS APPS →
07 · FAQ

Kibana and Arc0, answered.

Q01

Can I use Kibana with Claude, ChatGPT or Cursor?

Yes. Connect Kibana to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Kibana actions you allow.

Q02

How do users connect Kibana?

Users add their Kibana api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

Q03

Which Kibana actions can my agent take?

47 in total: 34 read, 6 write and 7 destructive, such as “create Case”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Kibana?

Yes. Actions like “delete List” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.

Q05

Can my own backend call Kibana too?

Yes. The same Kibana connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Kibana into your agent.

Your users connect Kibana once, under your brand. Your agent gets 47 actions behind your policies, with every call on the record.

Connect KibanaBrowse all apps →

Free to build · MCP + REST · Audit log on every plan