S/Runtime
Talk to an architect
Abstract enterprise intelligence control plane with luminous agent paths

Security and governance

Apply identity and policy at retrieval and action time.

Check identity, permissions, data policy, model policy, and action scope during the run, then retain the evidence required for review.

Talk to an architect Platform overview
Capability statusPublic architecture and product previewExact availability, provider support, and control ownership are confirmed in a signed pilot scope
Identity
Permissions
Privacy
Policy
Execution
Audit
Identity

Anchor AI access to the enterprise identity plane.

S/Runtime is designed to resolve SSO identity, groups, roles, source permissions, and workspace context before governed retrieval or action.

  • Map users to source-system access and team scopes.
  • Support role-based controls for builders, reviewers, admins, and consumers.
  • Configure permission checks across the AI surfaces included in the deployment scope.
Privacy

Control sensitive data before it enters prompts or tools.

Privacy controls identify, mask, minimize, or block sensitive content based on data type, region, role, and workflow risk.

  • Apply PII, PHI, financial, and contractual policies by source and field.
  • Support retention controls and context expiration.
  • Show policy explanations for blocked or redacted responses.
Audit

Make AI work defensible for regulated enterprises.

Consequential retrieval, model routes, agent steps, approvals, outputs, and actions are designed to retain policy and source evidence within the agreed scope.

  • Export audit trails for compliance review.
  • Measure agent risk, drift, and outcome quality over time.
  • Give administrators control over models, tools, connectors, and data scopes.

Set the limits before the agent runs.

Choose the permitted models, connectors, data scopes, tools, approval owners, and evidence requirements for this workflow.

Judge the system by the work it completes.

Pilot teams can test whether retrieval and action stay within the intended user and source boundary.

Security reviewers can define the evidence required to reconstruct consequential retrieval and execution paths.

Shared policy patterns can be scoped across search, assistants, APIs, and agent workflows.

Data access

Resolve the current user and source ACLs before retrieval, limit fields by policy, and preserve lineage for the records and documents included in a run.

Open page

Agents

Package an objective, permitted context, approved tools, policy checks, and review steps into a workflow with a named owner.

Open page

Platform

Connect approved data sources, check policy at each step, route sensitive writes for review, and keep the record needed to reconstruct a run.

Open page

Architecture session

Bring one consequential workflow. Leave with a governed agent blueprint.

We will map the context, systems, decisions, controls, actions, and success measures together.

Talk to an architect