IT Operations (ITOps) has thus become a critical success factor. Stability, scalability and cost control depend on one fundamental question: do you have control over your IT landscape?
That grip should come from the Configuration Management Database (CMDB). In practice, we only see that many CMDBs are filled, but hardly function as a control tool. Data is present, but insufficiently accessible and too polluted to base daily operational decisions.
This article describes how the Blis Digital CMDB Agent reduces that gap and how the CMDB is thus evolving from a static data source to a foundation for modern ITOps and the first step towards agentic, AI-driven operations.
The current challenge: CMDB data without impact
Almost every medium to large organization has a CMDB. Nevertheless, we see recurring problems in practice.
- The available data is complex and difficult to understand
- Access requires specialist knowledge or complex query languages
- Information is spread across multiple fields and sources
- Use is limited to a small group of CMDB specialists
As a result, valuable configuration and asset data is not being used enough. During incidents, insight is quickly lacking, changes are carried out with a limited impact overview and cost optimization remains fragmentary. The CMDB exists but does not function as an active control tool within iTops.
ITOps requires context, speed and accessibility
Modern ITOps include much more than just operational management. It affects, among other things:
- Incident, problem and change management
- Cost control and FinOps
- Compliance, Audits and Lifecycle Management
- Storage, archiving and data management
- Cloud and Platform Operations
All these disciplines need context: which systems are involved? In which environment? Who is the owner? What is the business impact? What are the costs?
The CMDB Agent: A New Way to Work
To bridge the gap between CMDB data and daily use, Blis Digital has developed the CMDB Agent. This agent is built on the Elastic platform and uses advanced AI-driven search models.
The CMDB Agent acts as a natural language interface on top of existing CMDB data and can be linked to a variety of source systems, such as call registration, changelogs, financial systems, audit data, and on-premise and cloud infrastructure.
Users no longer need to know data models or query languages. Questions like “Which production systems are part of application X?” or “Provide an overview of servers by environment” are answered immediately. This makes the CMDB practically useful for a broad target group: service desk, engineers, operations, FinOps and management.
Why Elastic is the right basis
Elastic is ideally suited as a foundation for a modern, AI-driven CMDB Agent because it combines scalable search technology, real-time analytics, and AI integration into one platform. Below, we explain the main reasons.
Search and analysis power at scale
Elastic makes it possible to quickly search over millions of configuration items. Users can freely filter for multiple attributes at the same time, such as environment, owner, type, and status. In addition, aggregations can be created directly, such as numbers, distributions and trends. This makes the platform highly suitable for interactive and natural language queries via the CMDB Agent.
Below is an example of such a natural language question:
“Can you give me an overview of Windows 2012 servers by environment?”
Schema-flexible: perfect for complex CMDB data
By definition, CMDB data is complex and changes continuously. Elastic is schedule flexible, making it easy to add new fields. Data from different CMDB sources can be brought together without a rigid data model that stands in the way of innovation. Performance remains stable even with very high data volumes, which is essential for CMDBs in complex hybrid and cloud environments.
Real-time insight without loss of performance
Elastic is designed for near real-time indexing and querying. New or modified configuration items can be searched almost immediately and aggregations remain fast, even with large datasets. As a result, the CMDB can be used 24/7 in operational processes, without being dependent on heavy batch processing.
Natural AI interface with control
Because Elastic was originally a search platform, it fits well with AI-driven interfaces. The translation from natural language into search intent, query and result is very robust and transparent. Elastic supports vector search, semantic search and relevance scoring, among others. In combination with Elastic Agent Builder, you can limit AI agents in their actions, and only approved ES|QL actions are executed. This way, answers can always be traced back to real data. This prevents hallucinations and increases confidence in the results.
One platform for CMDB, ITOps, and FinOps
Elastic is not a standalone search tool. It is a complete observability and analytics platform. As a result, the CMDB Agent can be seamlessly linked to logs, metrics and traces, incident and event data, and cost and consumption information. Here, the CMDB acts as a context layer for incident analysis, change impact and FinOps and cost optimization, without creating new data silos.
Security, governance and control are built in
For enterprise ITOps, governance is essential. Elastic offers:
- Role and Rights Structures (RBAC)
- Strict data access by index and field
- Audit logging
- Defined agent actions via Agent Builder
As a result, an AI-driven CMDB can be used safely in production environments.
Data sovereignty as a strategic advantage
CMDB data provides insight into the entire IT landscape and therefore requires maximum control. Elastic offers complete control over where your data is:
This means:
- Data stays inside own data centers or chosen regions
- No mandatory data export to external AI or SaaS platforms
- Full control over data residency
Essential for organizations with GDPR, AVG, NIS2, DORA or sectoral regulations requirements.
First, get the basics in order
An important starting point of the CMDB Agent is accessible value creation. In the first phase, the solution consciously focuses on:
- Insight into assets and configuration items
- Ownership and Team Responsibility
- Environments (DEV/ACC/PRD)
- System status, type, and location
This makes it possible to get a grip on the basics quickly, without major changes to existing processes or tooling. The CMDB Agent reads data but does not change it, making adoption safe and manageable.
The CMDB as a central context layer within ITOps
Within a mature ITOps architecture, the CMDB plays a central role as a context and connection layer between ITSM (incidents, changes and problems), observability and monitoring, FinOps and cost reports, security, audits and compliance, and lifecycle, storage and archiving processes.
The CMDB Agent makes this context directly accessible. This creates a shared view of the IT landscape, leading to better decision-making, faster escalations and more predictable changes.
AI-first and the step towards Agentic ITops
At Blis Digital, we work completely AI-first. In our own development and operation process, and in how we help customers move forward. The CMDB Agent is not an end point in this, but a conscious first step. A foundation on which we will continue to build towards Agentic ITOps.
In this vision, several specialized AI agents will eventually work together, each with a clear role and defined responsibility. Examples include a configuration and dependencies agent (CMDB), an agent that analyzes incident history and helps with root cause analysis, an agent that provides insight into costs, consumption and optimization from FinOps, and agents that support compliance, licensing, and audits.
The strength lies not in one smart agent, but in teamwork. Allowing these agents to work together creates an ITOps model in which analysis, correlation and decision-making are increasingly automated.
The CMDB remains the stable core here. The reliable source of context and structure that all agents fall back on. No black box, but a controlled, explainable foundation for AI-driven operations.
Technical Setup: What's Required
To implement the CMDB Agent, Blis Digital offers:
- Flexible infrastructure (cloud or on-premises deployment)
- AI Agent Configuration
- Secure integration with your data sources
- Ongoing management and optimization
- Timeframe: Proof of Concept: 2-4 weeks
- Deployment in production: 6-12 weeks (depending on data source complexity)
During deployment, there is no disruption to the existing configuration; the CMDB agent reads data but does not change it.
Ready to take the first step?
With the CMDB Agent:
- Get a quick grip on your IT landscape
- Bring your ITOps and FinOps together
- Lay the foundation for Agentic, AI-driven operations
Our proposal:
- Schedule a demo with your own data
- Start a 4-week Proof of Concept
- Work with us to develop a roadmap to autonomous ITOps
About the Authors
Cees Writing is an Account Manager at Blis Digital and has the main goal to provide clear added value to its customers in terms of services around the Elastic Stack. These services focus on Search&AI, Observability and Cloud Optimization. His idea, as with this CMDB agent, is that delivered solutions can be incorporated into the customer's larger ITops and/or FinOps framework, using Elasticsearch as the search engine that achieves the best results fastest, regardless of the amount of data, source systems or platform.
Etiosa Raymond is an Elastic Specialist at Blis Digital and focuses on building intelligent observability and search solutions using Elasticsearch, AI agents, and cloud-native technologies. He specializes in transforming complex data landscapes into accessible, automated systems that create value for the business.
