Last updated: 2026-05-24 04:00 ICT

Source: Robert KB + Git

Version: v1.1

Release: v3.1.0

4-Layer Architecture

A layered approach to AI orchestration that separates strategic intent, routing, execution, and truth storage.

1

Executive Layer

Strategic intent, decision authority, scope control

Robert / GPTHermes + Opus + custom MCP*

Hermes operates as a context-aware operator surface. MCP is the read-only bridge for context retrieval — not an autonomous execution path.

Strategic intent + scope
2

Routing Layer

Capability-based routing, worker allocation

optimize-worker
Routed task + capability match
3

Execution Layer

AI workers, policy guardrails, testing & critique loops

CodexHermeslocal agents
Reviewed artifact (commit gate)

Human Review Gate

Checkpoint, not a layer — every artifact passes here before it is treated as committed truth.

Committed truth
4

Source-of-Truth Layer

Committed KB (glossary + index), decision logs

Privacy-firstCommitted KB protocol

*Current operational trial surface. MCP is used as a read-only context bridge for controlled context retrieval. It is not yet approved as an autonomous production workflow.

Cross-cutting: Backlog & Visibility surface — control surface across every layer

E

AIOS Observability Layer

Captures execution traces, routing decisions, cost/usage evidence, reliability signals, and deployment readiness checks. In Phase 1, telemetry production and store behavior remain implemented in optimize-worker while AIOS owns the governance meaning of the evidence.

Governance evidencePhase 1 implementation

Surface

ai-os-profile remains the deployable cockpit and app surface.

Foundation

AWS foundation is ready, but the app is not deployed there.

Gate

Deployment waits for the SQLite runtime contract; refactor work is paused pending regression strategy.

Six Design Choices

1
Optimization & Routing

Match capability to task

2
Policy & Guardrails

Boundaries before behavior

3
Backlog & Visibility

Control surface, not admin

4
Knowledge Base Governance

Truth lives in committed knowledge

5
Testing Discipline

Critique, regression, gates at every layer

6
Human Review Gate

Last line before truth commits

“AI orchestration is program management. Tools change. Governance discipline doesn't.”

Component Glossary

ComponentDefinition
Robert KB + GitSource of truth. Master archive of decisions.
Robert / GPTExecutive orchestrator and reviewer.
Hermes + Opus + custom MCP*Context-aware operator surface — Active trial — structured context pull, task continuity, governance-aware operation.
optimize-workerExecution routing and task packaging layer. In Phase 1, it also implements telemetry production and store behavior for observability evidence.
CodexBounded implementation executor for repo edits.
ai-os-profileCurated external visibility layer and deployable cockpit/app surface (this site).
AIOS Observability LayerGovernance evidence layer for traces, routing decisions, cost/usage evidence, reliability signals, and deployment readiness checks. Not a claim that AWS app deployment is complete.

*Current operational trial surface. MCP is used as a read-only context bridge for controlled context retrieval. It is not yet approved as an autonomous production workflow.