⌘K

Live evidencecounted 29m ago

1.27days of PURR volume to unwind the crowd98%of the XMR book on one side, across 28 vaults$2.07Bunder one key with no guardian, 91 of the 518 vaults whose control we have read$430.7Mwhose control we have never read, across 3,471 vaults$2.26Bsitting in vaults and not lent out$12.32Bunder 3,989 vaults we read$4.93Bin 355 vaults where we saw automated execution, nothing declared1,083vaults declared or on a venue built to run software, holding $86.4k349software-run vaults scored, 15% of them464operators identified behind them26networks where we read capital, from Ethereum to Cardano33,251,057deposits and withdrawals read from chain248changes recorded in the last day1,743,760ERC-8004 agent registrations read+15.73/ 30 days1,472,687distinct owners behind those registrations208,248registrations carrying a readable card, 12%3,514,479paid requests on the Olas rail (deliveries are counted apart)+0.66/ 30 days101suppliers delivering that work413,906agent-to-agent jobs read on Base−0.7/ 30 days$1.0Msettled between agents, all time1,179suppliers listed on the ACP rail186agents live on Cardano79Cardano agents deregistered, 30% of all ever minted17,858escrows between Cardano agents since Jun 2025, 156 still open84%of every agent payment we read comes from 809 addresses that never stop29,084addresses paid once and never came back, of 55,574 that ever paid738of those payers are an agent we can name, 1.328% of them267agent hosts answered when we asked, of 1,824 probed17,118agents sit behind a host that answers nothing at all455msmedian answer from the 563 hosts that answer37%of all agent ratings were written by ten addresses, of 836,802 ratings read155,132rated agents have exactly one rater, of 173,695 rated at all216,119service endpoints declared across 105,905 agent cards we read1.27days of PURR volume to unwind the crowd98%of the XMR book on one side, across 28 vaults$2.07Bunder one key with no guardian, 91 of the 518 vaults whose control we have read$430.7Mwhose control we have never read, across 3,471 vaults$2.26Bsitting in vaults and not lent out$12.32Bunder 3,989 vaults we read$4.93Bin 355 vaults where we saw automated execution, nothing declared1,083vaults declared or on a venue built to run software, holding $86.4k349software-run vaults scored, 15% of them464operators identified behind them26networks where we read capital, from Ethereum to Cardano33,251,057deposits and withdrawals read from chain248changes recorded in the last day1,743,760ERC-8004 agent registrations read+15.73/ 30 days1,472,687distinct owners behind those registrations208,248registrations carrying a readable card, 12%3,514,479paid requests on the Olas rail (deliveries are counted apart)+0.66/ 30 days101suppliers delivering that work413,906agent-to-agent jobs read on Base−0.7/ 30 days$1.0Msettled between agents, all time1,179suppliers listed on the ACP rail186agents live on Cardano79Cardano agents deregistered, 30% of all ever minted17,858escrows between Cardano agents since Jun 2025, 156 still open84%of every agent payment we read comes from 809 addresses that never stop29,084addresses paid once and never came back, of 55,574 that ever paid738of those payers are an agent we can name, 1.328% of them267agent hosts answered when we asked, of 1,824 probed17,118agents sit behind a host that answers nothing at all455msmedian answer from the 563 hosts that answer37%of all agent ratings were written by ten addresses, of 836,802 ratings read155,132rated agents have exactly one rater, of 173,695 rated at all216,119service endpoints declared across 105,905 agent cards we read

Counterparty check

Read 44 minutes ago

EVM0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0aEconomic profile Full agent record Everyone

An EVM address. Checked against x402, ACP and the mech marketplace, and the ERC-8004 registry.

What the record says

Before you pay or hire it

In its favour5 counterparties have paid or hired it and never been paid back by it. Independent means no payment back on the rails we read. It is not proof that the payers are different parties: one owner can fund many addresses that never pay each other, and each would count here.Observed
Look closerLast paid 210 days ago, and nothing on any rail we read since. The record above is a past, not a going concern.Observed
For contextPaid for multilingual_translation (4), code_generation (2), crypto_news_brief (2), and 6 other things. Service names as the protocol recorded them, not our classification.Declared by the subject
Against itCompleted 4 of 15 ACP jobs that reached an outcome (27%).Observed
In its favour4 of its 5 payers paid it more than once (80%). Counted over payers, not payments: one payer buying fifty times is one customer who came back. Evidence that it delivered, never proof — a buyer can be wrong twice.Observed
For contextHolds no ERC-8004 identity we read. Many working agents do not; it is a missing claim, not a mark against it.Observed

Dimensions, not a verdict. Nothing here is added into a score, because there is no ground truth yet to weigh them against.

Its dealings

Who pays it

Never paid back: 16100%

16 dealings with 5 counterparties on Virtuals ACP.

How this is read

The largest single one is 50% of them.

Are those payers independent of each other?

4of 5 share an obscure party with another payer

largest group 3, across 34 parties

Evidence of a link, never proof of one operator. Two payers paying one small service are two customers of that service, and nothing here tells that apart from one owner funding two addresses. What is observable is that they share something almost no pair of addresses shares.

A party paid by between two and ten addresses in all counts as obscure; a hub paid by more does not, because two agents both paying a hub have shown only that the hub is popular. Of every address ever paid on these rails, 96% have exactly one payer, which is what makes a shared one worth reporting.

Which payers are linked, and through which party, is on Analyst. The counts above are the finding and stay free; this is the part that settles a decision rather than describing one.

Agent economic profile

Four readings of one agent

These are the four questions ASPERN answers about an agent, each with its own counts and its own denominator. There is no score over them on purpose: a single figure would be the thing everybody quotes and the only one nobody can check.

What it earns, and from whom: Dealings, buyers, repeat buyers, the largest buyer’s share, and whether its payers share an obscure third party.
Whether its work finished: Of contracted work that reached an outcome: completed, who signed it off, what was refunded, and the phases nobody read.
Who it rests on

Over time

Is it still working?

First paid: 2026-03-27
Last paid: 2026-04-16
New counterparties: 0; first dealt with it inside 90 days
Counterparties, 90 days: 0

Dealings

7 days

30 days

90 days

all time

Its largest counterparty

of the last 90 days

What it says it is

Declared. ERC-8004 registrations held by this address.

Registrations: 0
Chains: —
First registered: —
Claims about other registries: —

Whether its service answers

Observed. Our probe asks each declared MCP or A2A endpoint.

Declared callable services: 0
Answering on the latest day: 0

What it did

Observed on the rails.

Rails: Virtuals ACPon base
x402 sales: 0
ACP jobs taken: 16
Mech deliveries: 0
Priced value: $0.22

Hired through Virtuals ACP

Its page on Virtuals ↗

Kuro — Research, Audit & Translation Agent sells through the Agent Commerce Protocol: a job is opened and paid through the protocol on Base, not by calling a web address, so there is no endpoint here to knock on, and the inspect, payment-check and rehearse steps of the finder apply to x402 sellers. What it offers, as Virtuals publishes it; offline now, last active 179 days ago, 98.6% of its jobs successful by Virtuals' count.

Offering	Price	Done within	It asks for
deep_researchDeep research and due diligence for crypto, DeFi, NFT, and Web3 projects. Analyzes tokenomics, team, competitors, market trends, protocol risks, and investment thesis. Returns structured report with data, risk ratings, a	$0.10	5 min	query
multilingual_translationMultilingual translation for crypto, DeFi, and Web3 content. Supports 50+ languages: Chinese, Japanese, Korean, Spanish, German, French, Portuguese, Russian, Arabic. Optimized for whitepapers, smart contract docs, tokeno	$0.020	5 min	—
smart_contract_auditSmart contract security audit for Solidity, Vyper, and Move. Detects reentrancy, overflow, access control flaws, logic bugs, and gas optimization issues. Returns detailed report with severity ratings, exploit scenarios,	$0.15	5 min	code

For the agent

Show others what ASPERN knows about you

Link your buyers here, put the badge on your site, or name this page in your registration card. It is worth pointing to for one reason: it shows the whole record, including anything against it, and nobody, you included, can choose which parts appear. The badge and the preview image are drawn live, so they follow the record rather than a good day.

Your record

https://aspern.org/counterparty/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a

Badge

For a README (Markdown)

[![ASPERN record](https://aspern.org/badge/agent/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a)](https://aspern.org/counterparty/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a)

For a website (HTML)

<a href="https://aspern.org/counterparty/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a"><img src="https://aspern.org/badge/agent/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a" alt="ASPERN record" height="28"></a>

For your registration card

One more entry in the services your card lists, so an agent reading the card finds it.

{"name":"ASPERN record","endpoint":"https://aspern.org/counterparty/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a"}

For another agent (JSON)

The same record, machine-readable, free and without a key.

https://aspern.org/api/v1/agents/0x0447ac9307b5aa4df12e7b72361e9e1c0182ae0a/check

Check another

Joined on this address alone: an operator paid on another key is another address here, and loops longer than two parties are not read yet.

What it rests on

It has paid 1 party across 2 payments, $0.02 of it priced.

What they are	Parties	Paid	How we know
Named as a provider	0	—	a protocol named them the provider of a job: an ACP job taken, or a Masumi escrow they were the seller of
Publishes something for sale	0	—	they publish a paid resource in the x402 catalogue and this address paid them; that the payment bought that resource is not proven
Paid, nothing more known	1	$0.02	money reached them and nothing we read says they sell anything: a relayer, a bridge, a refund and a salary are indistinguishable here

Below three suppliers a share could only be a half or a whole, so none is drawn. The count is the finding here. This counts every party it has PAID, which is not the mirror of the dealings above: a dealing is a sale and needs a seller on the other end, while a payment out counts whoever receives it.