S.E.N.T.R.I.

The AI Assistant of ServerEngine.

Agentic Automation
Multi-Model Support
Skills & Approvals
Secure Architecture
icon
S.E.N.T.R.I.
Server Engine Neural Task Runner & Intelligence
LIVE!
Ollama, local gemma4 OpenRouter, gwen3-coder-plus OpenRouter, gwen3-coder-plus Claude, claude-sonnet-4-6 Claude, claude-sonnet-4-6 OpenRouter, gwen3-coder-next OpenRouter, gemini-3-flash-preview ServerEngine, Runbook Creator
S.E.N.T.R.I.

How it Works

1

You Command

Type an instruction to accomplish:
like “run iso compl. on dc1”

2

S.E.N.T.R.I Orchestrates

S.E.N.T.R.I interprets intent, calls required skills and orchestrates it to the automation layer if needed.

3

ServerEngine Executes

Skills with “.runbook” run in ServerEngine’s automation engine.

Designed for Secure Agentic Automation

S.E.N.T.R.I. only knows what hosts and skills are available — zero visibility into script logic, credentials, or execution. Authentication and execution live entirely in the automation layer.

Step 01
Interaction Layer
“run iso compliance check on dc1”
intent
Step 02
Orchestration Layer
S.E.N.T.R.I decides skill selection based on intent No script content No credentials
Trust boundary
Step 03
Automation Layer
WinRM authentication & execution
Secure credential retrieval
Automatic logging
Pass result back to S.E.N.T.R.I
Why this is how agentic automation should work
No credential access
AI never sees passwords, API keys, or service tokens. Authentication is handled exclusively inside ServerEngine’s automation layer.
Trust only visibility
The AI sees a skill’s name and description and host targets — nothing more. Script names and content stay behind the trust boundary.
Human-in-the-loop
Skills marked Approval Required pause before execution. S.E.N.T.R.I. waits for explicit human confirmation before any sensitive operation runs.