Module architecture

At a glance

Ontologie is organized into interconnected modules. Each module serves a specific function, and the modules communicate with each other to deliver a unified experience. This page explains how they fit together.

Overview

                    ┌──────────────┐
                    │  Dashboard   │
                    │  (Global view)│
                    └──────┬───────┘
                           │ KPIs from each module
        ┌──────────────────┼──────────────────────┐
        │                  │                      │
┌───────▼──────┐  ┌────────▼────────┐  ┌─────────▼────────┐
│  Ontologie   │  │ Workflow Builder│  │  Agent Studio    │
│ (Business    │  │ (Automation)    │  │  (Conversational AI)
│  model)      │  │                 │  │                  │
└───────┬──────┘  └────────┬────────┘  └────────┬─────────┘
        │                  │                     │
        │    ┌─────────────┼─────────────┐       │
        │    │             │             │       │
┌───────▼────▼──┐  ┌──────▼──────┐  ┌───▼───────▼───┐
│  Spreadsheets │  │  Calendar   │  │  Knowledge     │
│ (Tabular      │  │  (Events)   │  │  Base          │
│  data)        │  │             │  │  (Documents)   │
└───────┬───────┘  └─────────────┘  └───────────────┘
        │
┌───────▼───────┐
│   Live Data   │
│ (External     │
│  sources)     │
└───────────────┘

How modules interconnect

Ontologie → Everything

The ontology is the core of the system. The entities and relationships you define are leveraged by all other modules:

Module	Ontology usage
Agent Studio	The agent queries entities to answer questions
Workflow Builder	Workflows can react to entity modifications
Spreadsheets	Entity-type columns link rows to ontology entities
Live Data	Synchronized data is linked to existing entities
Dashboard	Ontologie KPIs reflect model completeness

Live Data → Spreadsheets → Ontologie

External data follows a natural path:

1. Live Data synchronizes data from your sources (Google Sheets, HubSpot, etc.)
2. The data appears in Spreadsheets (Synchronized data tab)
3. Rows are linked to Ontologie entities via entity mapping

Agent Studio → Ontologie + Knowledge Base

The AI agent combines two data sources:

• Ontologie: structured data (entities, properties, relationships)
• Knowledge Base: unstructured data (documents, web pages)

The agent uses a hybrid search to find the most relevant information from both sources.

Workflow Builder → All modules

The Workflow Builder orchestrates actions across modules:

• Triggers: entity events, schedules, webhooks
• Actions: create entities, send emails, call the AI agent, read/write spreadsheets
• Integrations: external API calls, synchronizations

Dashboard → All modules

The Dashboard is a read-only view that aggregates metrics from each module:

Section	Source
Ontologie KPIs	Number of entities, relationships, completeness
Workflow KPIs	Executions, success rate, average duration
API KPIs	Availability, performance, active endpoints
Agent Hub KPIs	Conversations, success rate, required actions
Timeline	Events from all modules

Governance → Ontologie + Data

Governance applies across the board:

• Classifications: sensitivity labels on entities and columns
• Approvals: validation workflows before modifications
• Quality: data quality rules
• Lineage: traceability of data origin

API and external integrations

All modules are accessible via the REST API and the MCP server:

Access	Usage
REST API	Programmatic integration (scripts, n8n, Make, Zapier)
SDK	TypeScript or Python client for your applications
MCP	Connection from Claude Desktop, VS Code, or any compatible AI client

For more details: API & Access.

See also

• Introduction — Platform overview
• Quick start (Interface) — First steps
• Quick start (API) — First API call

Need help?

Contact us: Support and contact.