Receipts, not renders.

Receipt 01

Pages on this site were shipped by the fleet.

The glossary entries human-in-the-loop and agent-sandbox were built by agents working inside Vivari - you're reading their output when you open those pages. The trail, end to end: NEXUS dispatched a researcher and a copywriter from one plain-language directive; the copywriter drafted into the shared knowledge vault; a human reviewed the draft and approved it; a terminal agent promoted it to code, verified the build, and committed - two commits, each with a written handoff note. Different runtimes did different jobs: the drafting and shipping ran on Claude, a copy-lint pass ran on Cursor.

Receipt 02

Autonomy is a setting a human chooses.

Mid-session, Vivari intercepted a tool request from an agent and asked its human how permissions should be handled from here - manual review of every request, auto-approval of safe actions only, or fuller autonomy. The work waited while a person decided how much rope the fleet gets; the moment they chose, the floor moved again. That's the management layer's quiet job: autonomy is a visible ladder someone climbed on purpose - never a silent default.

Receipt 03

The stranger asked to push to main. The house said no.

We gave a demo agent one small task whose next step was git push origin main - a step outside its permission profile. The risk evaluator held the call mid-flight: the popup you see is Vivari's own escalation gate, the notification center filed it as urgent, and the device itself rang. The push never ran. The deny landed in the audit trail - five times, actually, because we re-fired the gate until the footage was clean, and the trail kept every one. The task was staged. The gate is not.

Receipt 04

The review layer said no to us.

We asked the cognitive pair to brainstorm launch content. The Brainstormer found that two concepts we referenced weren't documented in its knowledge vault - and instead of inventing citations, it flagged the gap. The Curator (a different model family, by design) then reviewed the drafts and ruled:

verdict: revise
cross_model_diverse: true
rationale: the drafts are sound, but the evidence base was
incomplete - fix the vault, then re-brainstorm.

A review layer that only ever approves is decoration. This one sent us back to do the work - the vault gap it flagged is now filled, and the revision cycle is queued. That's the anti-slop machinery we ship, being applied to ourselves first.

The floor itself

Four harnesses, one dock.

The workspace the fleet ran in: real Claude, Codex, and Cursor terminals side by side - different models, one floor, one set of rules. Product footage from our demo reel, labeled as such per our own policy below.

Standing

Every deploy of this site is scored by Vivari Guard.

Before changes ship, the same deterministic engine that reviews agent changes scores this site's diffs against its own git history. Latest deploy:

risk 0.0/10 · confidence High · gate PASS · 2026-07-04

Recent diffs have been low-risk content additions, and the scores say exactly that - no drama to show you yet. When Guard holds something real on this repo, the receipt will be here.

The ledger

How access actually works.

Founding cohorts form this fall, around ten teams at a time. The constraint is real and operational: every workspace gets guided setup and calibration against your repositories, and that has a human ceiling per cohort. The request list is never capped - invitations go out in order of fit, not order of signup.

Our commitments on this page and everywhere else:

• No countdown timers. No fake scarcity. Capacity limits come with their operational reason attached.
• No invented numbers. Metrics appear here when they're real and meaningful, not before.
• Receipts are unstaged. Where footage is a recreation or a demo, it will say so.
• What we hold back, we hold back deliberately - early product internals stay ours.

Field note

Building this way cuts both ways: on day one of the fleet program, the fleet surfaced real bugs in our own tooling. They're being fixed, and honestly - that's the point. A workspace that manages agents has to survive its own agents first.