Skribe — Law. DONE. | FairPlane — Claim. DONE. | UNOY — Work. DONE.
UNOY Knowledge · For IT & Governance

A workforce with clear permissions.

Review how UNOY organises data access, integrations, changes and approvals for your use case. Get a basis for domain and technical decisions.

Roles, API, SSO, separate environments for build, staging and production, records of work and defined model usage form the technical framework. Which building blocks your deployment uses is agreed together.

Find your starting point → Prefer to have it done? Service
Have it done · Have it tailored · Shape it yourself
UNOY OS as a controlled operating architecture: roles, environments, audit trail
Have it done · Have it tailored · Shape it yourself

"Business units want to deploy AI, but we need controlled architecture rather than ungoverned agents."

§ 03 · Shared foundation

One foundation for different tasks.

Claim Workforce, Advisor and Workforce Basis use shared technical capabilities. Roles, controls and integrations are defined for each offer. Peppol Services is in development and pilot.

Entry gate · every new AI matter passes through here

01

New matter in governance

IntakeRisk-RatingInventory

New matter in governance.

Business unit submits AI initiative → risk rating, vendor due diligence and data classification → entry in AI usage inventory with approval, monitoring and escalation path.

02

Roles & Permissions

OwnerEditorViewerRWA

Roles & Permissions.

Employee identity → mapped to Owner / Editor / Viewer / User → access per workflow, table and field set at granular level.

03

SSO & Identity

IdPKeycloak · EntraRole · Department

SSO & Identity.

Existing identity provider (Keycloak, WKIS-SSO, Microsoft Entra) → authentication with context (role, department) → pre-filled workflow start without login friction.

04

API & Integrationen

OSCRMERPDMSAPI

API & Integrations.

Existing systems (M365, Google Workspace, S3, ERP) → OAuth2 or webhooks, REST or GraphQL → data flows are agreed per deployment, without parallel shadow databases.

05

Build / Staging / Production

BuildStageProductionTest → Approval → Rollback

Build / Staging / Production.

Workflow draft in Build → testing in Staging → go-live after approval, with rollback provided.

06

Audit trail in every run

Lauf #245114:32Skill v3 · Model14:30Approval · Role14:29

Audit trail in every run.

Workflow trigger → record of intake, skill version, model, output and actor → traceable storage for internal and external reviews.

Technology · Knowledge Base

Understand the technical building blocks.

The Knowledge Base explains structure and use. Verify available functions through your configuration and a suitable evaluation.

21

documented building blocks

EU

data hosting

ISO 27001 + 42001

Certificates of data centre operators

§ 06 · Operational Governance

Introduce changes through a controlled process.

AI can help improve workflows and skills. Tests, permissions and approvals govern production use. Progress and results are returned to the directing AI and responsible people.

01 · AI Usage Inventory

Which matters run through UNOY, with which models, data sources and business units. The overview can be exported within the agreed scope.

02 · Risk Ratings

A risk assessment per matter: data categories, scope of affected parties, impact of errors. Criteria and classification are agreed per deployment.

03 · Vendor Due Diligence

Sub-processor list, AI provider mapping, data locations, operator certificates. Versioned and provided within the agreed scope.

04 · Approvals & Monitoring

Who approved which matter, under what conditions, and in what state. Monitoring dashboard with status per workflow.

Legacy System Integration

Identity & Access

Azure AD, Okta, SAML, OIDC, SCIM

Mail & Office

Outlook, MS 365, Exchange, S/MIME

DMS & CLM

SharePoint, iManage, DocuSign

SIEM & Logs

Splunk, Datadog, Sentinel, Audit-Export

Certificates of the data centre operators used: ISO 27001 · SOC 2 Type II · ISO/IEC 42001 (AI Management). Data locations and transfers are defined for your deployment.

§ 08 · Next step

Which architecture question should we answer first?

Data locations, DPA, sub-processors, roles, API, records of work. Bring your checklist and we will go through it together.

Find your starting point → Discuss your task → Explore Self →

Have it done · Have it tailored · Shape it yourself