
Explain suspicious files and unusual behavior
using observed evidence
Behavioral Analyzer records process creation, file and registry changes, and network activity observed in an isolated sandbox. It scores each deviation from the UEBA baseline for the relevant user, server, service account, or AI agent. Integrated EDR or gateway tools handle blocking.
Deployment consultationWe’ll confirm the scope and deployment environment during the consultation. Deployment guide
Record suspicious file behavior
in an isolated environment
Run email attachments, newly introduced packages, and scripts of unknown origin in a VM disconnected from external networks.
Record events such as shadow copy deletion, mass encryption, and connections to domains in chronological order. Apply YARA detection rules to files created during execution.
- Record the process tree and file, registry, and network events
- Record execution conditions such as time and random inputs so results can be compared
- Check for detection gaps caused by delayed execution or differences between environments
Activity that interferes with recovery+3.1s
Entropy 7.9+6.4s
Domain registered 3 days ago+7.0s
Persistence after reboot+7.2s
How far does it deviate from normal?
See a breakdown by category
Behavioral Analyzer learns separate baselines for sign-in times, download volume, repositories accessed, and upload destination domains. Each alert shows what deviated and by how much, so analysts can see why it fired. Early baselines are less certain when data is limited. Operations teams review alerts and exceptions to refine them.
- Separate baselines for each user, server, and service account
- Use Okta and Entra ID sign-in logs and EDR telemetry
- Manually register exceptions such as leave, department transfers, or financial close periods.
Apply the same checks
to AI agent tool calls
Build a baseline from the tools agents call, how often they call them, and which destinations they reach.
Flags unusual outbound messages or repeated queries as detection signals.
- Collect tool call logs from MCP servers and API gateways
- Detect outbound sends, bulk queries, and out-of-scope tool calls
- The integrated MCP gateway enforces blocking
Normal rangeToday
Normal rangeToday
Deviation score: 0.9414:22
12 times the baseline14:22
Key features
Isolated sandbox
Windows and Linux executables, Office documents, PDFs, and PowerShell and shell scripts
UEBA
Baseline each account using its own activity patterns
Agent behavior analysis
Flags unusual sequences or frequencies of tool calls
Network activity
Periodic C2 beacons, DGA domains, and attempted SMB or RDP lateral movement
Reproducible execution
Record execution conditions for comparison during review
Isolate generated files
Files generated in the sandbox stay within the analysis network
How it works
Collection
Suspicious files, EDR telemetry, and IdP and proxy logs
Isolated execution
Run in the sandbox and log observed behavior
Compare against the baseline
Calculate deviation score for each item
Classification
Map behavior to ATT&CK techniques and record the supporting evidence
Handoff
Send results to Threat Correlator
Integration tools and methods
Integrations use APIs and syslog. We’ll confirm what each tool supports based on its version and your operating environment.
Deployment options
| Method | Description | Recommended |
|---|---|---|
| On-premises and isolated networks | The platform runs in your data center. Offline packages deliver detection rule and model updates so the platform can operate without an external connection. | Finance, public sector, and defense |
| Dedicated cloud | A dedicated environment serves one customer in a South Korean cloud region. | Enterprises and healthcare |
| Managed | In this MDR option, Genox handles security monitoring. We’ll confirm the scope and response procedures during the consultation. | Small and medium-sized businesses and mid-market companies |
| MSSP partners | The security service provider operates separate tenants for each customer. | Security service providers |
Frequently asked questions
Can sandbox results vary between runs?
Repeat the run with matching conditions where possible, including time, random inputs, and network responses, then compare the logs. Some behavior may be absent in isolation if a file depends on external responses or reacts to its execution environment.
Could this be used to monitor employees?
Baselines use only metadata permitted by policy, such as access time, transfer volume, and accessed resources. Customers decide what to analyze and how long to retain it. Before deployment, review any notice requirements under the Personal Information Protection Act with your legal and HR teams.

Discuss a Behavioral Analyzer deployment
Tell us about your operating environment, and we’ll explain what the demo and proof of concept can cover.