Caution: delivery not yet confirmed
Trust assessment for this x402 endpoint, updated every monitoring cycle. All of this seller's endpoints · Run this endpoint? Get it tested
https://xhagents.xyz/api/arkham-intel/counterparties
- Score
- 74 of 100
- Uptime, 7 days
- 100%
- Median response
- 1192 ms
- Test payments
- Not yet paid-tested: priced above our automatic test budget ($0.05 per call), so it stays C (unconfirmed) until a funded test confirms delivery (see below)
- Payout-wallet incidents
- 0
- Current price
- 0.1 USDC
- Browser agents
- Can't pay it yet: CORS headers missing (server-side agents are fine)
- x402 version
- v2
For the seller: make it payable from browser agents
- The 402 response has no Access-Control-Allow-Origin header, so a browser can't read the price quote. Fix: send Access-Control-Allow-Origin: * on 402 responses.
- The quote is in the PAYMENT-REQUIRED header, but it isn't listed in Access-Control-Expose-Headers, so browser code can't see it. Fix: Access-Control-Expose-Headers: PAYMENT-REQUIRED, PAYMENT-RESPONSE.
Checked 2026-11-06. This doesn't affect the grade.
Based on 196 checks over the last 7 days. How scores are made
Run this endpoint? Get it tested
This endpoint costs more per call than our tester pays on its own ($0.05), so it isn't paid-tested automatically. Verify your endpoint: you fund one test at your own price, paid to our tester wallet and held for that test. Our tester pays your endpoint that same amount (so it goes back to your payout wallet), anything the test doesn't spend is returned to you, and the grade comes only from what the test finds. The test runs at a random time within about 3 hours, and the result is posted on the GitHub issue automatically. If it can't run within 3 days, the funding is refunded. You can fund another test whenever you like (one waiting at a time), and every funded test and its result is listed on this page.
Seller guide: payout wallets, known-answer tests and more. Changed your paycheck.json? Check my setup.
Recent paid tests
No paid tests in our recent history. Delivery isn't confirmed yet, so the grade is C at best.
Each paid test sends your listing's own example input (or your declared test request). Fixed something? Use the Get it tested box above.
Paid tests since Nov 7, 2026 come with a signed receipt: the exact payment terms and response (as hashes), amount, result and time, signed with our published key. How to check a receipt · All receipts for this endpoint (JSON)
Recent failed checks
The free checks we run every 30 minutes: uptime and a valid payment quote. Paid tests are in the section above.
1 check didn't pass in the last 7 days, and none of them count against this endpoint. It was on our side (our monitor being rate-limited, blocked, or having a problem), not the endpoint.
| Time (UTC) | What happened | What it means | Affects score? |
|---|---|---|---|
| 2026-11-03 20:56:33 | Monitor problem | Our monitor had a network problem during this check. Not the endpoint's fault. | No |
What each result means
- Blocked
- The endpoint's protection (such as Cloudflare) rate-limited or challenged our monitor. That's about our monitor, not the endpoint, so it never affects the score.
- Request mismatch
- The endpoint rejected the format of our test request (HTTP 400, 405, 415, or 422) before quoting a price. That's about our request, so it never affects the score.
- Network not supported yet
- The endpoint asked for payment on a network our monitor can't read yet. That's our limitation, so it never affects the score.
- Monitor problem
- Our own network had an issue during that check. Never affects the score.
- No response
- The endpoint didn't answer within 10 seconds, even after a retry. Counts as downtime.
- No payment quote
- The endpoint answered, but not with the x402 payment request it's listed with. Counts as downtime.
- Unreadable quote
- It asked for payment, but the payment details were malformed, so an agent couldn't pay it. Counts as downtime.
Think this grade is wrong? Dispute it on GitHub. The form is pre-filled with this endpoint; a bot replies with its current score and queues a paid re-test.
We check every endpoint every 30 minutes from Render (US East), user agent lumiere-paycheck-prober/1.0 (+https://lumierepaycheck.org). Operators can allowlist it. Raw data: /v1/failures
Real usage
Payments that actually settled into this endpoint's payout wallet over the last 30 days.
Settled volume
$4.67
Payments
50
Distinct buyers
10
Top buyer: 71% of volume
Typical payment
$0.08
Largest payment
$1.00
Last 7 days
▼ 39%
vs the week before
Settled by 0xb87e…5860 (30%), 0x8f5c…e531 (11%), 0xa32c…d101 (9%) · first seen 2026-09-25.
This wallet receives payments for 34 monitored endpoints, so the totals cover all of them, not just this one.
Counted from public USDC payments on Base and Solana submitted by recognized x402 facilitators, the services that settle x402 payments. It's a floor: payments on other networks, or before our index started, aren't counted. Daily data: /v1/usage?url=…
Payout-wallet changes
No payout-wallet changes in the last 7 days.
Every change is checked against the seller's own declaration, the seller's earlier wallets, and direct transfers between the old and new wallet on Base. Confirmed rotations don't affect the grade; a change nobody can confirm is "avoid" for 24 hours, then "caution". Run this endpoint? Publish your wallets at https://xhagents.xyz/.well-known/paycheck.json as {"payTo": ["0x…"]} so future rotations are recognized instantly. Raw data: /v1/wallet-changes?url=…
Buyer reports
No buyer reports in the last 7 days.
Agents that pay through Lumière PayCheck report whether the response was usable, one vote per buyer. Reports don't change the grade by themselves: problem reports from several buyers trigger our own paid re-test, and only that can change the score.
Run this endpoint?
Show buyers and their agents that your endpoint is monitored. The badge updates on its own.