Kibana MCP. Manage your entire observability stack via natural language.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Kibana. Manage your entire observability stack directly from your AI agent. This server lets you list, create, and update spaces, dashboards, and index patterns across your Elastic environment.
You can copy objects between spaces, get full object metadata, or automate the lifecycle of your monitoring setup using natural language.
What your AI agents can do
Add case comment
Adds a comment to an existing support case.
Bulk create saved objects
Creates multiple saved objects (like dashboards) in a single batch operation.
Bulk get saved objects
Retrieves details for multiple saved objects at once.
Create, delete, update, or list entire Kibana spaces to segment dashboards and visualizations for different teams.
Create, read, update, delete, or copy dashboards, index patterns, and visualizations across your instance.
Create, find, enable, disable, or update alerting rules to manage your system monitoring alerts.
Get or update data views, and create runtime fields to process and structure raw log data.
List and retrieve details for Kibana roles, allowing you to audit permissions and access controls.
Execute specific connector actions to interact with external data sources.
Ask AI about this MCP
Supported MCP Clients
Waiting for input…
Kibana MCP Server: 55 Tools for Data Observability
These tools let you programmatically control every aspect of your Kibana instance, from creating spaces to managing complex alerting rules.
019e5d2aadd case comment
Adds a comment to an existing support case.
019e5d2abulk create saved objects
Creates multiple saved objects (like dashboards) in a single batch operation.
019e5d2abulk get saved objects
Retrieves details for multiple saved objects at once.
019e5d2abulk update saved objects
Updates several saved objects simultaneously.
019e5d2acopy saved objects
Copies saved objects, such as dashboards, from one Kibana space to another.
019e5d2acreate agent policy
Creates a new policy for Elastic Agents.
019e5d2acreate case
Creates a new support case record.
019e5d2acreate connector
Creates a new data connector for Kibana.
019e5d2acreate data view
Creates a new data view or updates an existing one.
019e5d2acreate enrollment key
Generates a new enrollment key for an Elastic Agent.
019e5d2acreate or update role
Creates or updates a specific Kibana user role and its permissions.
019e5d2acreate rule
Creates a new alerting rule based on defined criteria.
019e5d2acreate runtime field
Adds or modifies a calculated field within a data view.
019e5d2acreate saved object
Creates a specific saved object, like a dashboard or visualization.
019e5d2acreate short url
Generates a short, shareable URL for a resource.
019e5d2acreate space
Creates a new, isolated Kibana space for a specific team or project.
019e5d2adelete cases
Deletes multiple support cases records.
019e5d2adelete connector
Deletes a specified data connector.
019e5d2adelete data view
Deletes an entire data view from the system.
019e5d2adelete role
Removes a Kibana user role and all associated permissions.
019e5d2adelete rule
Deletes an alerting rule that was previously created.
019e5d2adelete saved object
Deletes a specific saved object like a dashboard or visualization.
019e5d2adelete short url
Removes a generated short URL.
019e5d2adelete space
Deletes an entire Kibana space.
019e5d2adisable rule
Temporarily disables an alerting rule without deleting it.
019e5d2aenable rule
Activates a previously disabled alerting rule.
019e5d2aexecute connector
Runs a configured data connector action to pull current data.
019e5d2aexport saved objects
Exports defined sets of saved objects into a usable file format.
019e5d2afind rules
Searches and lists all existing alerting rules.
019e5d2afind saved objects
Searches the entire instance for saved objects matching specific criteria.
019e5d2aget agent
Retrieves detailed information about a specific Elastic Agent.
019e5d2aget case
Retrieves the full details of a specific support case.
019e5d2aget connector
Fetches the configuration details for a specified data connector.
019e5d2aget data view
Retrieves the full configuration and metadata for a data view.
019e5d2aget role
Fetches the details of a specific Kibana user role.
019e5d2aget rule
Retrieves all details for a specific alerting rule.
019e5d2aget saved object
Retrieves the full metadata for a specific saved object.
019e5d2aget short url
Retrieves the details and status of a short URL.
019e5d2aget space
Retrieves the full details of a specific Kibana space.
019e5d2aimport saved objects
Imports a set of saved objects from a specified file upload.
019e5d2alist agent policies
Lists all configured policies for Elastic Agents.
019e5d2alist agents
Retrieves a list of all currently enrolled Elastic Agents.
019e5d2alist connectors
Lists every available data connector configured in the system.
019e5d2alist data views
Retrieves a list of all existing data views.
019e5d2alist enrollment keys
Lists all available enrollment keys for agents.
019e5d2alist roles
Lists all user roles defined in Kibana.
019e5d2alist spaces
Retrieves a list of all available Kibana spaces.
019e5d2aresolve import errors
Checks and resolves errors encountered during the bulk import of saved objects.
019e5d2asearch cases
Searches the support case database for specific records.
019e5d2aunenroll agent
Removes a specific Elastic Agent from the system.
019e5d2aupdate cases
Modifies the details of existing support cases.
019e5d2aupdate data view
Updates the configuration or settings of an existing data view.
019e5d2aupdate rule
Modifies the parameters or logic of an existing alerting rule.
019e5d2aupdate saved object
Updates the metadata and settings of a specific saved object.
019e5d2aupdate space
Modifies the settings of an existing Kibana space.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Kibana, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Kibana MCP Server - Manage Dashboards & Observability
Your AI agent lets you control your entire observability stack right from the prompt. This server gives you the tools to manage everything in your Elastic environment—from spaces to dashboards and rules. You'll use your agent to list, create, update, and delete resources, letting you manage your data lifecycle using plain language.
Managing Kibana Spaces and Roles
You can segment your dashboards and visualizations by creating, deleting, or updating entire Kibana spaces using create_space, delete_space, and update_space. You can also manage user access by listing and getting details on roles with list_roles and get_role, or by creating or updating specific permissions with create_or_update_role and delete_role.
Handling Saved Objects
Need to manage dashboards or visualizations? You can create a new saved object, like a dashboard, with create_saved_object. You can also bulk-create or bulk-update multiple saved objects at once using bulk_create_saved_objects and bulk_update_saved_objects. To keep your environments consistent, copy saved objects from one Kibana space to another using copy_saved_objects. You can search for existing objects using find_saved_objects, or get the full details of any object with get_saved_object.
You can also delete specific objects with delete_saved_object, or export groups of objects into a file with export_saved_objects.
Alerting and Monitoring Rules
To keep tabs on your system, you can create a new alerting rule using create_rule, or update an existing one with update_rule. You can find all existing rules with find_rules, and you can temporarily disable a rule with disable_rule or activate it with enable_rule. If a rule is garbage, you can delete it entirely using delete_rule.
Data Views and Connectors
When it comes to data, you can create a new data view or modify an existing one using create_data_view or update_data_view. You can also enhance your data views by adding or changing a calculated field with create_runtime_field. You'll need to run external data sources using execute_connector after creating or fetching connector details with create_connector or get_connector.
You can list all available connectors with list_connectors and remove one with delete_connector.
Agents and Support
For your monitoring agents, you can list all enrolled agents with list_agents, or get details for a specific agent with get_agent. You can also generate new enrollment keys with create_enrollment_key and manage agent policies using list_agent_policies and create_agent_policy. If an agent is retired, you can remove it from the system with unenroll_agent.
For support issues, you can create a new case with create_case, search existing ones using search_cases, or get full details on a case with get_case. You can also update details on an existing case with update_cases or modify a case's comment with add_case_comment. Finally, you can delete multiple cases with delete_cases or modify a case's details with update_cases.
How Kibana MCP Works
- 1 Subscribe to the Kibana MCP Server and provide your Kibana URL and API Key.
- 2 Your AI agent calls a tool (e.g.,
create_space) using natural language. - 3 The server executes the action against your Kibana instance and returns the resulting object or status.
The bottom line is: your AI agent performs complex Kibana actions without you having to navigate the UI.
Who Is Kibana MCP For?
This is for the DevOps Engineer who needs to move a dashboard from staging to production without manual clicks. It's for the Data Analyst who needs to find a specific index pattern across a massive environment. If you manage observability data or dashboards, you need this.
Audits space configurations, moves dashboards between environments, and automates the provisioning of monitoring stack components.
Manages the lifecycle of monitoring rules and spaces, ensuring consistency between production and staging setups.
Finds specific visualizations or index patterns using natural language search across a complex Kibana instance.
What Changes When You Connect
- Move dashboards between environments instantly. Use
copy_saved_objectsto move an object from staging to production without manual steps. - Audit permissions and structure easily. List all roles with
list_rolesand get full details for a role usingget_roleto check who can see what. - Automate monitoring rules. Instead of manually clicking through rule editors, use
create_ruleorupdate_ruleto set up alerts based on natural language input. - Maintain data consistency. Use
create_data_vieworupdate_data_viewto standardize how raw logs are interpreted across multiple monitoring dashboards. - Find anything fast. If you don't know the exact ID, use
find_saved_objectsto search for dashboards or index patterns using plain text queries. - Control the data lifecycle. Use
delete_spaceordelete_saved_objectto clean up old, unused monitoring components and reduce clutter.
Real-World Use Cases
Staging-to-Production Dashboard Promotion
A DevOps engineer needs to promote a newly built dashboard. Instead of manually exporting and re-importing, they tell their agent: 'Copy the dashboard 'Global Network View' from the 'Staging' space to the 'Production' space.' The agent uses copy_saved_objects and confirms the promotion.
Standardizing Data Collection
A data analyst joins a new team and needs to see metrics from a legacy system. They ask their agent to 'Create a data view for the legacy logs and set the runtime field for latency.' The agent runs create_data_view and create_runtime_field, ensuring the new data is structured correctly immediately.
Mass Cleanup of Old Assets
An SRE is tasked with cleaning up old monitoring environments. They prompt the agent: 'List all spaces that haven't been used in 90 days, and then delete the saved objects inside them.' The agent runs list_spaces and then systematically uses delete_saved_object across the identified spaces.
Initial Setup and Permissions Check
A platform engineer needs to provision a new team workspace. They ask the agent to 'Create a new Kibana space called 'Project Phoenix' and grant the 'Developer' role read-only access.' The agent executes create_space and create_or_update_role.
The Tradeoffs
Searching for specific objects
Typing 'Find me the index pattern for logs' and hoping the AI understands the exact API call needed. This fails because the AI needs specific identifiers.
→
Use find_saved_objects. This tool lets you search the entire instance for saved objects using natural language queries, giving you the exact ID needed for subsequent commands like get_saved_object.
Manual configuration updates
Going into the UI to update a dashboard's time field or change a role's permission, wasting 15 minutes on clicks.
→
Use update_saved_object or update_role. Tell your agent exactly what needs changing. For example: 'Update the 'User Dashboard' saved object to use the time field '@timestamp'.'
Guessing object IDs
Trying to run delete_saved_object with an ID you remember but which is slightly wrong, resulting in a 404 error and broken workflow.
→
First, run get_saved_object to confirm the object's existence and metadata. If you need to search for it first, use find_saved_objects.
When It Fits, When It Doesn't
Use this server if your job involves managing the full lifecycle of monitoring data—creating, moving, modifying, or deleting spaces, dashboards, and rules. You need this if your workflow requires consistency across development, staging, and production environments. Don't use it if you only need to view data (that's a simple query). Don't use it if you only need to manage user authentication (use a dedicated identity management tool instead). This server is for structural changes to the observability stack itself.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Kibana. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 55 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Navigating a sprawling Kibana UI is a time sink.
Right now, moving a dashboard requires multiple clicks: you find the dashboard, you select 'Save As,' you choose a new space, you confirm the overwrite, and you repeat this process for every single visualization you want to promote. It's tedious, and you're always worried about missing a step or losing the correct version.
With this MCP server, you just tell your agent: 'Copy the 'API Latency Overview' dashboard from 'Staging' to 'Production'.' It uses `copy_saved_objects` and handles the entire transfer, giving you the final, correct dashboard in seconds.
Kibana MCP Server: Manage Data Views & Dashboards
Manual data preparation involves checking multiple tabs to see which fields are available, then manually defining the data type and transformation logic for every new metric. You have to confirm that the field is available before you can use it in a dashboard.
Now, you tell your agent to 'Create a data view for the logs and add a runtime field to calculate the request latency.' The agent runs `create_data_view` and `create_runtime_field`, setting up the field and the calculation in one go.
Common Questions About Kibana MCP
How do I use the Kibana MCP Server to copy a dashboard? +
You use the copy_saved_objects tool. Simply ask your agent to copy the object, specifying the source and destination spaces. You don't need to know the underlying API IDs; just give the names.
Can I use Kibana MCP Server to list all available spaces? +
Yes, run the list_spaces tool. It returns a full list of every space in your instance, allowing you to see exactly where your dashboards live.
What is the best tool for finding dashboards in Kibana MCP Server? +
Use find_saved_objects. This tool lets you search the entire instance for saved objects using plain text queries, making it much faster than navigating the UI.
How do I manage user permissions with the Kibana MCP Server? +
You use the create_or_update_role or list_roles tools. These let you create new roles or update existing ones, managing who can see what data.
How do I delete a space using the Kibana MCP Server? +
Use the delete_space tool. This permanently removes the entire space and all its contained objects. Be sure you know what you're deleting before running it.
How do I use the `get_saved_object` tool to check the metadata of a specific index pattern? +
The get_saved_object tool retrieves full metadata for any specific saved object. You just need to provide the object ID and type to see its configuration, last updated time, and field mappings.
What should I do if I run into errors when I try to `bulk_update_saved_objects`? +
If bulk updates fail, check the object IDs and ensure the data structure matches the target saved object type. The API response will detail which records failed and why.
Does the Kibana MCP Server support listing all Kibana roles using the `list_roles` tool? +
Yes, the list_roles tool lists every Kibana role in your instance. This lets your agent review all available roles to understand user permissions and access levels.
Can I search for a specific dashboard across my entire Kibana instance? +
Yes. Use the find_saved_objects tool and specify the type as 'dashboard'. You can also provide a search string to filter the results by name or description.
Is it possible to move dashboards from a development space to a production space? +
Absolutely. The copy_saved_objects tool allows you to select objects from a source space and replicate them into one or more target spaces, including their references.
Can I create new Kibana spaces using this integration? +
Yes, you can use the create_space action. You just need to provide the space configuration JSON, including the ID and name you want for the new environment.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Currents
Access real-time global news and search millions of articles directly from your AI agent using the Currents API.
Appbot
Analyze app reviews and sentiment with Appbot — track user feedback, ratings, and topics across iOS and Android via AI.
Harvard ClinicalTrials
Search and analyze clinical trial data from Harvard and ClinicalTrials.gov for research, drug development, and healthcare innovation.
You might also like
Veraset
Equip your agent to seamlessly query Veraset's mobility datasets. Run geospatial SQL, extract insights, and manage S3 buckets.
CoinLore
Access real-time cryptocurrency data, market stats, exchange info, and social metrics directly from any AI agent.
Bleez
Power your Brazilian e-commerce with a platform that integrates payments, inventory, and logistics for the local market.