Start here

How it works

1 · It looks at your product

Sage loads the product in a real browser rather than reading the HTML. It follows links, opens what a visitor can open, and records the states it can actually reach — screenshots, rendered content, console errors, forms it could fill. The output is a map of the product as a visitor experiences it, not as the marketing copy describes it.

It reads only public pages by default. It never signs in or spends anything unless you hand it a dedicated test account, and even then it uses a mailbox it owns rather than asking you for an inbox credential.

2 · It writes the missions

Missions come out of what it saw. If it found a signup flow, there is a mission about the signup flow; if it never reached one, there is not. A deterministic gate sits between the model and the plan — a mission that cannot be verified, or that asks a tester to do something the agent could not confirm exists, is discarded before you are shown it.

You review the plan once and approve it. That is the first of the two human moments. See Missions & budget for how rewards are derived.

3 · Real people complete them

Funded campaigns appear on a public board. Testers pick a mission, do the work, and submit a written account of what they saw. There is no application process and no interview — the work is the application.

4 · It checks the work against its own browsing

This is the part that makes a payout mean something. A tester’s account has to line up with what Sage itself observed in the product, so a report that could have been written without opening it does not clear. Any quote in a decision must be an exact substring of the evidence; anything else is dropped. See Judging evidence.

5 · It pays, or it refuses

If the work clears, the vault computes the reward and settles USDC on-chain, and a public receipt is published citing the evidence it paid for. If it does not clear, the submission is refused with a written reason — and the reason says plainly when the refusal is not a judgement on the tester’s work, which is often the case when a campaign has simply ended or filled.

6 · Then it decides again

Under a standing mandate the loop closes on itself. Sage reads what the board produced — what was claimed, what was ignored, what it cost — and proposes the next piece of work to buy: which surface, what kind, with its reason and its price, before anything is spent. The price is never the model’s; it comes from the ceilings the founder set. See Fund once, then stop deciding.

The same loop pays for anything you can define

Testing is one instance of a more general machine: verify a claim against evidence the agent gathers itself, then move money inside hard limits. So the same five steps run gigs (“pay my designer $50 when the logo page is live”) and milestone grants (“split J$10,000 across two milestones”) and supplier payments (“release $60 when the delivery confirmation for invoice INV-1042 is public” — trade finance whose paperwork is the payment itself) — composed by hand on /launch, compiled deterministically with no model in the path, and priced in the founder’s own currency: the Caribbean denominations are first-class, converted once at a stamped, source-attributed rate, settled in USDC.

And the loop leaves something behind: every payout lands on the earner’s verified work record, which the Advance facility can lend against — repaid by the waterfall from their next verified payouts. Money out, a record built, capital back in.

Where the human actually is

Exactly twice in the whole lifecycle: approve the plan, and fund it. Under a standing mandate, once: fund your account, and stop any move you do not like inside its window. Everything after that the agent does on its own and narrates afterwards with an artifact backing each claim.