DataTugPull more from your data
Get started
☰

Structure · meaning · instances · evidence

Help people and AI
make sense of data.

Data is easier to use when its shape, meaning and source stay connected. ModelSpec describes the model, MeaningGraph adds shared meaning, and OVDB Directory helps locate concrete database instances. DataTug starts from your actual data: deterministic schema analysis, AI-assisted investigation, and built-in inventory and environment management help build that knowledge from the ground up. Connecting these pieces into a shared knowledge graph is the direction we are building; MeaningGraph and Directory API integrations are planned.

See the connection

Four pieces, connected

The graph we are building.

Each layer has a distinct job. Together they help people and AI find relevant data and understand how to work with it.

M01

ModelSpec · structure

A storage-neutral description of entities, fields, types and relationships. It says what exists in a model, independently of where it is stored.

↗02

MeaningGraph · shared meaning

Will link model elements to concepts and definitions people can reuse, keeping meanings connected to the fields they explain.

D03

OVDB Directory · instances

Helps you find published database instances and inspect their models and deployment details: where a logical model meets a concrete source.

T04

DataTug · inventory & evidence

Deterministic schema analysis and AI-assisted investigation help build knowledge from your data, alongside built-in inventory and environment management.

One model. Every environment.

Know what it is.
Find where it lives.

ModelSpec defines the logical model regardless of where or how it is stored. OVDB Directory helps you find a specific published instance and inspect the model and deployment it describes.

DataTug has data inventory and environment management built in. Keep DEV, QA, UAT and PROD sources organized with their environment context, so you can understand which instance you are investigating and compare environments against the same model.

The public Directory is browsable today. Connecting its API to DataTug’s inventory is planned; the diagram is an illustrative inventory, with no live connections.

Browse OVDB Directory
MODELSPEC · LOGICAL MODELInvoiceSame definition, different instances
OVDB DirectoryInstance identity · model · deployment
  • DEVDevelopment instance
  • QAQuality assurance instance
  • UATUser acceptance instance
  • PRODProduction instance
DataTug inventorySources organized by environment

Illustrative relationship · not a live environment comparison

Follow one field

From “BillingCountry” to a country you can identify.

A database field may contain a name, abbreviation or internal code. A connected meaning could help people see the intended concept and where a mapping came from. DataTug integration is planned.

Illustrative path — a mapping becomes dependable only after its source and meaning are checked.

MODEL

What this will enable for people and AI

Less guessing. Better next steps.

01

Find relevant sources

People and AI will be able to start from a business concept and find related modeled fields and data sources.

02

Understand relationships

A person or agent will be able to see why datasets might connect, then inspect a proposed join before using it.

03

Make choices with evidence

The graph is being designed to keep definitions, mappings, source details and verification status visible to people and agents.

The workflow we are building

Discover. Build. Verify. Reuse.

DataTug helps build knowledge from the ground up as well as use it. Start with deterministic schema analysis: inspect entities, fields, types, keys and relationships, and organize concrete instances in the inventory. Use AI-assisted investigation to propose meanings and mappings, then review them against real sources.

That evidence is the starting point for ModelSpec models, MeaningGraph definitions and OVDB instance records. The shared graph connections and Directory API integration are planned; proposals become trusted knowledge through verification.

  1. DiscoverAnalyze schemas and inventory instances
  2. BuildShape models, meanings and instance records with AI assistance
  3. VerifyInspect provenance and real values
  4. ReuseShare approved understanding

Incidents reveal. People connect.

Nobody knows everything.
Build understanding together.

Incidents are moments when hidden knowledge and edge cases surface: an unexpected value, a relationship nobody documented, or a difference between QA and PROD. Investigating them reveals how a system really behaves.

Collaboration is the key. A developer knows the schema, an operator knows the environment, a domain expert knows what a value means, and AI can help propose connections. Bring those perspectives together, check the evidence, and keep what the team confirms.

DataTug helps investigate the data. Incidentius keeps the incident, its context and response. The knowledge graph we are building will connect reviewed discoveries back to the incident that taught them, so the next person or agent can use that understanding.

DataTug integration with Incidentius: Planned.

Keep what incidents teach · Incidentius
THE MOMENT OF DISCOVERYAn incident exposes an edge caseUnexpected data · hidden assumptions · environment differences
  • DeveloperStructure and relationships
  • OperatorInstances and environments
  • Domain expertMeaning and exceptions
  • AI assistanceCandidate explanations to check
Review together. Keep the evidence.Shared knowledge with its source and incident context

Illustrative workflow · Incidentius integration is planned

Shared concepts. Your domain.

Reuse common meanings, keep your own context.

Shared definitions can help across projects. Your domain-specific models and mappings remain yours to shape, with their scope and provenance visible.

DataTug integration with MeaningGraph: Planned.

ModelSpec is an early preview; generators are not available. MeaningGraph connects definitions to fields, while DataTug’s deeper integration is still planned. The graph illustration describes the direction and the pieces; availability is shown in the product.

Explore the product

Bring structure and meaning into your workflow.

Ways to use DataTug →AI and DataTug ↗