Run in your environment.
Agents read your alerts where they are and send out only verdicts and references.
Trust
Your security data never reaches SOCGenX.
Our agents run inside your environment. This page shows what crosses the boundary, how the agents are kept in check and which independent assurance we plan.
Data boundary
What reaches SOCGenX, and what never does.
The control plane receives metadata and verdicts. Your logs, alerts and evidence stay where they are.
| Data | Reaches SOCGenX? | Where it stays |
|---|---|---|
| Raw logs and telemetry | Never reaches SOCGenX | In your SIEM and log stores. |
| Alert contents and evidence files | Never reaches SOCGenX | Read by the agents inside your environment. |
| Credentials for your SIEM and tools | Never reaches SOCGenX | Held in your environment and never sent to us. |
| Personal data inside your logs | Never reaches SOCGenX | Stays with the logs it belongs to. |
| Detection metadata, such as rule names and status | Reaches SOCGenX | Sent to the control plane. |
| Verdicts the agents reach | Reaches SOCGenX | Sent to the control plane. |
| Evidence references, not the evidence itself | Reaches SOCGenX | Pointers back into your systems. |
| Health signals for each agreed log source | Reaches SOCGenX | Sent to the control plane. |
| Contact details you send through our form | Reaches SOCGenX | Sent by email; see the privacy notice. |
AI safety
How the agents are kept in check.
The agents work inside limits your team sets, and each limit is visible in the Evidence Ledger.
Recommend first.
By default agents read and recommend. They do not change anything in your environment.
Approval before action.
A response runs only after an analyst approves it, or when it belongs to a low-risk class your team pre-authorised.
Evidence before verdicts.
An agent must show the evidence behind a verdict before it can recommend an action.
Kill switch.
Your team can pause every agent at once and drop them back to read-only.
Full audit trail.
Prompts, tool calls, decisions and approvals are recorded in the Evidence Ledger.
Assurance roadmap
Independent assurance we plan to pursue.
SOCGenX does not hold any of these today. Each one stays marked as planned until an independent auditor confirms it.
Planned SOC 2 Type II
An independent auditor's report on how our controls operate over time.
Planned ISO 27001
An information security management system assessed against the standard.
Planned Cyber Essentials Plus
A hands-on technical check of our baseline security controls.
We will update this page when a status changes. Until then, ask for the security pack for detail on our current controls.
Security pack
What your security review will ask for.
Ask for the pack and we will send it by email.
- Architecture and data-flow overview
- The data boundary in detail
- AI guardrails and the autonomy ladder
- Answers to your security questionnaire
We reply from info@socgenx.com.
Review the boundary on your own stack.
We walk your security team through the data boundary and the guardrails, using your SIEM.