Skip to main content

Customer Journeys

These journeys show how customers move through Keeptrusts once access is available. Use them when you want a task-oriented map instead of a page-by-page product tour.

Inspect individual runtime requests with kt events tail or GET /v1/events with its required since query. The console Overview summarizes aggregate activity; it does not replace request-level event lookup. History adds session context only for capture-enabled flows.

Journey map​

Getting live​

JourneyStarts hereSuccess signal
Evaluate quicklyQuickstartThe client succeeds and the event query returns the matching request ID
Run a connected gatewayInstall the GatewayA connected runtime is healthy in Gateways
Set up team ownershipOnboarding TeamsAccess, review, and billing owners are clear

Operate and improve​

JourneyStarts hereSuccess signal
Roll out a policy changeConfigurationsNew version is deployed and behaves as intended
Investigate an unexpected decisionkt eventsRoot cause and next action are clear
Review human-review workInboxReview item is claimed, resolved, rejected, or returned to the shared queue
Export evidence for a reviewTrailEvidence package is ready for handoff

Govern spend​

JourneyStarts hereSuccess signal
Review spend and budgetsUsageThe team can explain recent spend and any budget actions

Journey 1: Evaluate quickly​

  1. Finish Quickstart.
  2. Send a representative governed request.
  3. Run kt events tail --since 10m --json and confirm the request ID and governance outcome.
  4. If conversation capture is enabled, confirm the request in History.
  5. Open Overview and confirm the organization is showing live activity.

Use this journey for evaluations, prototypes, and first-time product walkthroughs.

Journey 2: Run a connected gateway​

  1. Install kt using Install the Gateway.
  2. Start the runtime with kt gateway run.
  3. Confirm the runtime appears in Gateways.
  4. Send a governed request and confirm its request ID and outcome with kt events tail --since 10m --json.
  5. If conversation capture is enabled for the flow, confirm the captured session in History.

Use this journey when your organization needs the runtime inside its own infrastructure perimeter.

Journey 3: Set up team ownership​

  1. Invite members and assign roles.
  2. Confirm who owns policy rollout, inbox review, audit exports, and billing controls.
  3. Verify each person can reach only the surfaces intended for their role.
  4. Review Onboarding Teams for the operating checklist.

Use this journey when moving from a single-user evaluation to a shared organization.

Journey 4: Roll out a policy change​

  1. Start from a template or an existing configuration.
  2. Validate the YAML locally.
  3. Save or review the change in Configurations.
  4. Deploy to the target gateway.
  5. Verify runtime state in Gateways.
  6. Confirm representative traffic with kt events tail --since 10m --json or GET /v1/events?since=10m. Check History only when conversation capture is enabled, and watch Inbox if review volume may change.

Use this journey for controlled rollout of policy, routing, or provider changes.

Journey 5: Investigate an unexpected decision​

  1. Query kt events tail --since <window> --json or GET /v1/events?since=<window> and find the affected request ID. If conversation capture is enabled, use History for the captured session context.
  2. Review verdict, policy results, provider details, and timestamps.
  3. Compare the request with the configuration version that was active.
  4. Open Trail if you need audit chronology or evidence export.
  5. Decide whether the result reflects intended policy, a tuning issue, or a broader incident.

Use this journey for day-to-day support, rollout follow-up, and incident triage.

Journey 6: Review human-review work​

  1. Open Inbox and filter to the relevant review category.
  2. Claim the item when you are actively working it.
  3. Review the linked request context.
  4. Cross-check the captured session in History when capture is enabled; otherwise use the linked event and request metadata.
  5. Resolve or reject the item, or unclaim it back to the shared queue.

Use this journey when Keeptrusts raises an escalation or approval task.

Journey 7: Export evidence for a review​

  1. Open Trail and switch to the evidence export view.
  2. Choose the time window and export format.
  3. Select Request export job and wait while the job is queued or processing.
  4. When the job reaches completed, download the export and confirm the date range and scope.
  5. Pair the file with the request IDs, incident ID, or review note that explains why it was generated.

Use this journey for audits, compliance reviews, customer investigations, and incident response.

Journey 8: Review spend and budgets​

  1. Open Usage and set the time window you need.
  2. Inspect the drivers of cost, requests, or tokens.
  3. Review Budgets if the question is about budget thresholds.
  4. Review Bills or Payments when the question is about invoices, funding, or payment readiness.

Use this journey when finance, ops, or engineering needs a shared view of AI spend.

Next steps​