Analyse
Identify risks and configuration gaps.
Autonomous IT Operations
Your IT operator for the agentic AI era.
OPERGON analyses your IT environment, proposes actions and executes them under your rules. Verifiable results, always under your control.
Scroll
to explore
The problem
They leave you a list of alerts. We deliver the fix, validated and proven.
Vendors expose your posture, then stop. Your teams are buried under Active Directory audits, account reviews, GPOs and compliance evidence. Remediation stays manual, slow and rarely proven.
NTLMv1 allowed on 3 serversCRITICAL
SPNs encrypted with RC4 (Kerberoasting)CRITICAL
Windows LAPS not deployedHIGH
Service accounts with no expiryHIGH
Orphaned GPOsINFO
NTLMv1 disabled · validated, executed, provenPROOF
How it works
From analysis to proof, in 4 steps.
IT operations executed under your rules. Verifiable results.
Propose
Prioritise actions and present the expected impact.
Execute
Carry out authorised actions and log every change.
Verify
Check the results and deliver proof of execution.
🔒 Validation under your rulesBetween “Propose” and “Execute”: nothing runs without your go-ahead.
OPERGON by the numbers
Operations under control. Verifiable results.
key steps
Analyse, propose, execute, prove.
control point
Requests, validations and tracking, centralised.
of actions traced
History and proof of execution.
unauthorised executions
Your rules. Your permissions. Your control.
AI acts. You stay in control.
The three action levels
You decide how far the agent goes.
Every operation declares its level. Moving up a level requires explicit validation.
01 · read-onlyRead-only
Audit, analysis, report. No change to your environment.
02 · draft changeDraft change
Prepares the GPO, script, ticket or plan. Nothing is applied.
03 · execute approvedExecute approved
Executes only after explicit validation, reversibly and with full logging.
The OPERGON Index
A score out of 1000. Reproducible, sourced, actionable.
Each control is linked to its sources and frameworks (MITRE ATT&CK, NIST, CIS, ISO 27001, SOC 2) and weighted by severity. The same directory gives the same score: you measure progress, not an impression.
642OPERGON Index / 1000
×10Critical
×6High
×3Medium
×1Low
Sample value
The Proof File
Every fix leaves evidence.
- 01Before / after, UTC time-stampedWith the exact diff of every changed attribute.
- 02Separate identitiesThe human validator and the executing agent are distinct and named.
- 03SHA-256 chained integrityEach entry seals the previous one.
- 04Logged rollbackUsable for NIST, ISO 27001, SOC 2, cyber insurance and internal audit.
proof-file · OP-NTLM-0142
timestamp 2026-10-05T09:52:14Z
operation NTLMv1 restriction · GPO "DC-Auth-Hardening"
level execute approved
- LmCompatibilityLevel = 3
+ LmCompatibilityLevel = 5
validator human · separate identity
executor OPERGON agent · dedicated account
sha256 9f2c…e41a ← 7b10…03cd
rollback available · logged
Validated by a humanSample
Born in the field
Designed by architects who ran identity & access management for large enterprises.
Over twenty years of Active Directory and Entra ID in banking, defence, energy, healthcare and industry. Every OPERGON playbook encodes a remediation actually carried out in production.
Banking and finance
Defence
Industry
Healthcare
Energy
Sovereignty
Hosted in Europe. No dependency outside the EU.
The platform is hosted in Europe, with no dependency on any cloud or AI outside the European Union. Secrets stay in your vault; the agent acts with least privilege.
HOSTED IN EUROPENIST-ALIGNEDGDPRSOC 2
Diagnostic