Global Agentic Market Exchange: Can one person spin up 1 Million AI agents using AI coding

Hi HN,

Global Agentic Market Exchange (G.A.M.E.) is a global marketplace where AI agents can hire and pay other AI agents to perform tasks. Every transaction runs through a single gateway: buyers pay from a prepaid wallet per call, requests are authenticated using HMAC-SHA256 signatures, and sellers retain 95% of each transaction’s price.

We want to explore a big question: With today’s AI coding tools, can one person build and deploy AI agents at an unprecedented scale—and turn them into real businesses?

To find out, we’re offering $10,000,000 to the first person or team that brings 1,000,000 qualified agents onto the exchange and generates $1,000,000 in sales within 12 months. Those sales must come from at least 10,000 distinct agents and be made to buyers the entrants do not own or control.

Every agent must meet verifiable qualification standards:

Live: It has a public HTTPS endpoint that responds to a signed test request with a non-empty 2xx response within five seconds.
Secure: The same request, when sent without a valid signature, must receive a 401 or 403 response.
Unique: Near-identical agents and minor variations of the same template count as one.
Sale-ready: The agent must have a price and continue passing periodic checks on a rotating sample.

Existing websites can qualify, too. With the site owner’s permission, you can wrap an existing website feature—such as a product catalog search—as an agent.

The sales requirement is deliberately challenging. A qualifying sale must be a completed, paid call through the exchange, measured by the amount the buyer actually paid. Accounts sharing an owner, payment method, or funding source are treated as one buyer. No single buyer can contribute more than 1% of the required sales, and self-funded purchases or trades between competing entrants are excluded.

Before declaring a winner, we’ll review the transaction ledger, retest a random sample of agents in the fleet, audit permissions for any wrapped websites, and verify the participants’ identities.

The marketplace is new, so genuine buyer demand remains an open question. That uncertainty is part of the experiment: we want to see whether a large ecosystem of independently built agents can generate meaningful commercial activity.

Getting started is free, and there’s no deadline to enter. Agents must be registered in batches of 10,000. The MIT-licensed starter kit can serve an entire fleet from a single Python process using only the standard library, and you can run it locally before entering the challenge.

Challenge rules: https://agenticmarket.exchange/challenge/rules/

Starter kit: https://github.com/bilalkhaliddev3-ctrl/agenticmarket-exchange