Skip to main content
Online
60 FPS
12% CPU
24ms
1 Conn
00:00 session
LIVEv2.1.0

AURA Lab

Multi-agent coordination in real time

Interactive lab for exploring AURA's multi-agent orchestration — spawn agents, inspect tool routing, and watch context windows fill in real time.

FocusExploring multi-agent orchestration (AURA MCP) alongside code-intelligence and governance experimentation (AURA Core Architect) — two distinct components under one identity.

agentsorchestrationmcpmemorygovernancecode analysismodel integrationobservabilityautomation
TypeScriptPython 3.12MCP ProtocolLangGraphNeo4j
Interactive Demo
AURA Lab · v2.1.0live
AURA / Orchestration Runtime
140ms avg47 tools2,847 req/today
Live Dispatch Feed
●Geminianalyze146ms
●Geminigenerate→web.fetch113ms
●Claudeclassify→code.exec121ms
●Geminisummarize88ms
Claude
complete
GPT-4o
idle
Gemini
complete
Execution Flows

AURA MCP — orchestration flow

  1. 1Request↓
  2. 2Orchestrator / intent router↓
  3. 3Agent selection↓
  4. 4Context / memory↓
  5. 5Tool / MCP execution↓
  6. 6Observation↓
  7. 7Next decision↓
  8. 8Result

AURA Core Architect — analysis flow

  1. 1Repository / file↓
  2. 2Ingestion↓
  3. 3Knowledge / memory↓
  4. 4Analysis↓
  5. 5Metrics↓
  6. 6Proposal↓
  7. 7Governance↓
  8. 8Human approval
Component Matrix
Component Repository Purpose Status Technology
Orchestration Core private source Agent coordination, intent routing, tool execution Active TypeScript · Node.js · MCP
Automation Connectors private source Make, n8n, Power Automate, Zapier integrations Active TypeScript
Code Architect private source Code analysis, memory and governance studio Development React · Supabase
Edge Functions private source 20 Supabase functions (chat, memory, analysis, ingestion, governance…) Development Supabase
Repository Relation
2 private source repositoriescorearchitect
aura-portfoliopublic portfolio
System Architecture
ClaudeAnthropic GPT-4OpenAI GeminiGoogle AURA RouterArbitration Memory GraphNeo4j MCP Tools47 tools Action LayerExecution
Tech Stack
TypeScriptruntime
Python 3.12runtime
MCP Protocolprotocol
LangGraphml
Neo4jdata
Redisdata
Dockerinfra
Kubernetesinfra
Development Roadmap
Roadmap
●Phase 1
Router Core
Multi-model routing
MCP tool integration
Basic memory layer
●Phase 2
Graph Memory
Neo4j persistent memory
Cross-session continuity
Entity resolution
◉Phase 3
Agent Mesh
A2A protocol support
Sub-agent spawning
Capability discovery
○Phase 4
Autonomy Layer
Goal-directed execution
Self-healing agents
Continuous learning loop
IALABORATORIA

A laboratory for AI and technology content.

Systems

  • AURA Orchestration
  • MCP Ecosystem
  • Graph Memory
  • AI Agents
  • Docker Infrastructure
  • Industrial Intelligence

System Status

PlatformOperational
APIHealthy
3D EngineActive
MCP Nodes8 Online

Try the Konami code...

Stay in the loop

Occasional updates on AI systems, autonomous infrastructure, and new releases.

© 2026 LaboratorIA. All systems operational.

RSSNext.js 16 + React 19 + R3F + TailwindCSS v4