See what governance
looks like in practice.
Each tour starts with a captured product surface from the real app, then uses an illustrative walkthrough to explain how the same task plays out with and without LumenFlow.
Your AI just booked a $400 flight.
Who approved that?
Sidekick handles everyday tasks — travel, expenses, scheduling. Without governance, you have no idea what it accessed or spent.
Real Sidekick chat surface with the consumer-first entry and the governed workspace shell.
Captured from the real app in a local seeded workspace
The task completes, but nobody can reconstruct what happened or why it was allowed.
14:02:01
searching flights LAX → JFK
14:02:08
found 3 options, selecting cheapest
14:02:12
accessing payment method ending 4829
14:02:14
booking confirmed — $412.00 charged
14:02:15
calendar event created
14:02:16
done
No approval. No budget check. No receipt.
Policy, approvals, and evidence stay attached to the work all the way through completion.
task/booking scope-boundary verified
budget-gate: $412 within $500 trip limit
payment $412 → owner approval required
@you approved via notification
receipt attached: booking + payment + calendar
Budget checked. You approved. Evidence yours.
Use the same scenario with Sidekick in public.
These public scenarios are now real anonymous Sidekick runs on a hosted OSS runtime sandbox: sandbox-only sample data, live runtime artifacts, and no external writes. Pick one that matches the motion you are touring right now.
Every AI action should be legible.
Start governing yours.
One workspace. Policy, approvals, and evidence from day one. No credit card required.