Build an Agentic AI framework on top of your data space
Data spaces let participants share data and services under a common Rulebook: governed, standardised, and trustworthy. The pressing question today is how to use AI without losing control of your data, either personal and non personal. Onecub answers with Agentic AI frameworks that are ‘data space native’: policy-aware, auditable, and interoperable across organisations and sectors.
What we deliver
AI Agent creation & orchestration
Design and orchestrate modular AI agents that execute discrete tasks, chain into end-to-end workflows across complex value chains, and securely interface with your data space assets under governed access.
Policy-in, data-out
agents execute within the data space’s Rulebook (purposes, scopes, retention), never bypassing governance.
Human-in-the-loop
Review and approval checkpoints for high-impact steps (e.g., sign/submit).
Fine-grained permissions
Consent, mandates and delegations (individuals & organisations) enforced at run-time.
Provenance & audit
Verifiable logs, usage control, and tamper-evident traces for compliance.
How it works (at a glance)
Agents
plan, call tools/APIs, process data with AI and non AI programmes (inc. LLMs), negotiate contracts, call other agents.
Access
is controlled by the data space connectors and policy engine; sensitive processing can run in data clean rooms.
Identity
and consent/mandates are verified through EUDI-compatible wallets (persons & organisations).
Key standards & building blocks
Data space interoperability
IDS Dataspace Protocol (DSP), Eclipse Dataspace Components (EDC), Prometheus-X connectors for personal data use cases.
Identity & trust
eIDAS 2 / EUDI Wallets, verifiable credentials, qualified e-signatures/e-seals.
Catalogues & semantics
DCAT for catalogues, pivot and standard vocabularies/ontologies, ODRL/SHACL policies.
Agent orchestration
Support for Model Context Protocol (MCP) and tool/plugin patterns.
Typical outcomes
Trustworthy AI
Agents only see what policy allows, with explainable decisions and full traceability.
Faster operations
Automated evidence collection, reconciliation, and document generation.
Lower integration cost
Reuse of data spaces and sector Rulebooks, connectors, and semantic assets.
Federated innovation
One framework, many data spaces.
Reference projects
Gen4Travel
Agentic AI framework for the EONA-X mobility & tourism data space (disruption handling, multimodal journeys, service coordination).
Gen4Legal
Agentic AI for the Legal Data Space and Compliance-X (obligation discovery, evidence gathering, machine-to-authority submissions).