Who actually paid
x402 payments on Base · 20 October 2026
In the last 24 h on Base, no x402 payment reached this seller’s wallet.
Payments to 2 hosts land in this same wallet — usually one operator, sometimes a platform collecting for several. The group’s page: x402stock.xyz.
Daily money · x402 USDC on Base · 14 Oct 2026 – 20 Oct 2026 · the wallet’s, across every host paid into it
14 Oct $1 (3 payers) · 15 Oct under $1 (3 payers) · 16 Oct $0 (0 payers) · 17 Oct under $1 (1 payer) · 18 Oct under $1 (2 payers) · 19 Oct under $1 (2 payers) · 20 Oct $0 (0 payers)
Relationships
x402 payments on Base · 13 October 2026 – 20 October 2026 · 8 days
16 buyers paid this wallet over x402 on 6 of the window’s 8 days: 88 payments, $1.93. 0 of 16 buyers came back on another day (0%).
In the full record: its most loyal buyers, what is bought alongside it, who left it for whom, who came to it from where, and its early buyers, each with the counts behind it. How to get it: the $0.001 check over x402, Atlas Pro at $49 a month and for the seller itself, the Your buyers report at $29 a month.
Relationships are patterns in the x402 payments we observe on Base over the window: a buyer is a wallet that paid this seller’s payTo wallet, and came back means it paid on two or more different days. A wallet is not an agent, and a relationship is a pattern of payments, not a claim about who anyone is or who runs a wallet.
Is this your service? See who your buyers are: every wallet that paid you, day by day, and where the ones that left went, in one report.
Endpoints
The busiest of this seller’s 99 endpoints in the registry, by paid calls over 30 days; price and chain per endpoint.
| endpoint | paid calls | payers | price | chain |
|---|
/api/v1/quote/:ticker | 25 | 4 | $0.01 | Base |
/api/v1/avg-interest-rates | 21 | 4 | $0.01 | Base |
/api/v1/congress-trades | 18 | 3 | $0.03 | Base |
/api/v1/report/:ticker | 8 | 1 | $0.5 | Base |
/api/v1/dividend-calendar | 6 | 2 | $0.02 | Base |
/api/v1/dossier/:ticker | 5 | 3 | $0.05 | Base |
/api/v1/treasury-yields | 5 | 4 | $0.01 | Base |
/api/v1/fed-funds | 4 | 4 | $0.01 | Base |
/api/v1/grouped-daily | 4 | 2 | $0.02 | Base |
/api/v1/research/:ticker | 4 | 1 | $0.75 | Base |
/api/v1/trending/reddit | 4 | 2 | $0.01 | Base |
/zcj/neis63n1al62/api/v1/avg-interest-rates | 4 | 4 | $0.01 | Base |
87 more endpoints not shown.
The replay
Paid calls, rolling 30 days · 13 October 2026 to 21 October 2026
2026-10-132026-10-21
Selling something like this
Matched automatically from each seller’s own short description. It will sometimes be wrong; tell us and we fix it.
| seller | paid calls | price | sells |
|---|
| x402stock.pay.zeroclick.io | 283 | $0.01–$0.02 | Get the latest available price for any US-listed stock: the most recent 1-minute bar close plus today's OHLCV, |
| nojunk-stocks.mick8511.workers.dev | 8 | $0.001–$0.003 | Stock Quotes quote: One symbol -> latest price, change, previous close, day high/low, volume, 52-week high/low |
| api.tickersfeed.net | 655 | $0.001–$0.08 | Real-time or 15-min delayed stock quote and share price for any US equity ticker — current price, open/high/lo |
| edgar.apitoll.cloud | 90 | $0.003–$0.006 | Latest SEC EDGAR filings for any US-listed company by stock ticker or CIK: 10-K annual reports, 10-Q quarterly |
| data.tollkit.dev | 70 | $0.002–$0.03 | SEC filings lookup: give a US stock ticker or CIK, get the company's recent SEC EDGAR filings (10-K, 10-Q, 8-K |
| api.marketintelligenceapi.com | 673 | $0.001–$5 | Crypto news and stock market news, latest headlines: crypto and US stock headlines plus material corporate fil |
| x402stock.xyz | 760 | $0.01–$0.25 | End-of-day OHLCV bars for every US-listed ticker on a single date; one call returns the entire market. Pass `? |
| api.hyperextend.xyz | 87 | $0.001–$0.1258 | The most-recent N OHLCV candles (≤5000). Same per-row pricing as the windowed read ($0.00001 base for main/spo |
Going rate, the median of the 15 matched sellers with a listed price (of 15 matched): $0.01 a call. This seller’s endpoints run $0.01–$0.75, and the going rate falls inside that range — so no single over/under verdict is fair. Compare per endpoint.