Govern the AI Workers you build, buy, or deploy — with customer-controlled identity, authority, human oversight, and evidence.

Passport Studio

Turn AI Worker intent into a reviewable operating record.

Passport Studio helps customer teams define who owns an AI Worker, why it exists, what it may access or do, which risks and reviews apply, what evidence is required, and when its authority must pause, change, or end.

One Governed Record

Review the identity, boundary, and evidence before the AI Worker acts.

A Passport brings the operating questions into one governed record instead of scattering them across intake forms, tool configurations, approval notes, and dashboards.

Identity and accountability

Record the AI Worker, business purpose, outcome, accountable owner, sponsor, reviewers, risk tier, and lifecycle state.

Operating boundary

Define allowed and prohibited actions, data classes, models, tools, APIs, connectors, destinations, autonomy, escalation, and pause conditions.

Review and evidence

Show required Human Review, Toll Gate posture, evidence gaps, monitoring needs, renewal conditions, and the next accountable decision.

Passport To Action

The Passport establishes eligibility; the Control Plane decides at action time.

Passport Studio prepares the governed identity and boundary that later controls evaluate. A Passport does not grant credentials, execute tools, or authorize every future action.

1. Define

Create the Passport identity, purpose, owner, risk tier, lifecycle, permissions, prohibitions, and evidence requirements.

2. Review

Route missing evidence, higher-risk boundaries, exceptions, and consequential use through Toll Gates and accountable Human Review.

3. Recheck

At action time, evaluate current Passport scope, policy, tool authorization, evidence, destination, approval, and revocation state.

4. Permit and broker

When eligible in an implemented customer environment, issue a short-lived Runtime Permit and mediate the exact action through the Action Broker.

5. Record

Write decisions, receipts, exceptions, outcomes, and lifecycle changes to the evidence trail for accountable review.

Passport Studio supports governance compilation and review preparation. It does not issue credentials, approve production use, operate live connectors, provide legal conclusions, provide compliance conclusions, validate security posture, or replace accountable customer review.

Worked Example

See why a request can stop even when the AI Worker exists.

A Procurement Intake AI Worker has an owner and an approved supplier-update purpose, but a requested external destination falls outside its Passport boundary.

Passport outcome: review required

The record keeps the owner, purpose, approved scope, prohibited external-send action, missing destination evidence, and Human Review route together.

PassportOwner and purpose found
ScopeDestination outside approved boundary
Toll GateBlocked until evidence exists
EvidenceHuman Review item and Stamp recorded

This is a static public example, not a customer record or proof of live enforcement.

Review the example read model

Protected Workspace

Customer records and decisions stay behind customer-controlled access.

Passport records may contain operational context, reviewer notes, system boundaries, evidence references, tool posture, and lifecycle signals. The public page explains the product; it does not expose the workspace.

Scaled Agents provides

The licensed control-plane package, Passport record model, review workflows, lifecycle controls, evidence structure, and customer-managed deployment path.

Customers control

Identity, access, infrastructure, data, connected systems, policies, reviewers, evidence, deployment, and formal legal, security, privacy, compliance, audit, and production decisions.

The public page proves

Only the published product narrative and static example behavior—not customer configuration, workspace access, live connector execution, or customer-specific control effectiveness.