Paywary

Leaderboard / api.craigmbrown.com / endpoint

POST https://api.craigmbrown.com/v1/services/agent.prehire-check

Scores measure reliability and honesty, not output quality.

64Cautionconfidence: medium
  • Answered 4 of 6 probes in the last 7 days (67%).
  • Answered 4 of 6 probes in the last 30 days (67%).
Category
Signal by llm, confidence 60%
Latest quote
— on — to —
Uptime
67% / 7d, 67% / 30d
Delivery
none not yet paid-tested
Observed
none reported yet (opt-in wrapper telemetry)
HTTPS
yes
First listed
2026-11-16
Last probe
10h ago
Next probe
2026-11-18 05:48 UTC

Score components

Uptime, 7 days
19w 15%
Uptime, 30 days
19w 15%
Valid x402 challenge
100w 15%
Price honesty
100w 15%
Delivery after payment
n/aexcluded
Wallet stability
100w 10%
Listing age
7w 5%
HTTPS
100w 5%

Catalog listings

SourceSchemeNetworkAssetListed pricePay toFlowCurrentLast seen
cdpexactPolygon0x3c49…33592500000x5E70…4EB9defaultyes6h ago
cdpexactBase0x8335…2913$0.250x5E70…4EB9defaultyes6h ago
cdpexacteip155:50420x3600…00002500000x5E70…4EB9defaultyes6h ago

Paid probes

No paid probes yet. Paid probing is off by default and runs weekly per host under a strict spend cap; each paid probe links to its on-chain transaction here.

Delivery checks

Inside each paid probe we ask for something whose correct answer we know, then compare. A pass means the seller delivered what it was paid for; a fail means it charged and returned something wrong; inconclusive means we could not judge (no reference, unknown input shape, or a non-2xx response). Every row carries the request, the response hash and the reference so the verdict can be reproduced. Judged results feed the "Delivery after payment" component (pass 1, error after payment 0.25, fail 0); inconclusive rows are excluded; a fail in the last 14 days caps the verdict at caution. Methodology.

No delivery checks yet. They run inside the weekly paid probe.

Classification history

Every classification is kept; the newest row is the current category. Rules match the listing text; the model is consulted only when the rules are unsure.

WhenCategoryConfidenceMethodVersionEvidence
14h agoSignal60%llmv1claude-haiku-5-5: The endpoint returns a signed pre-hire due-diligence report on an AI agent (settlement history, dispute rate, flagged incidents), which is a risk verdict style report rather than a raw data lookup. It is closest to signal (risk verdicts and AI-enriched reports), though compliance or enrichment could be argued since it screens a counterparty.
30h agoOther40%rulesv1rule none matched "" in none

Probe history

WhenKindStatusLatencyx402ValidQuotedPay toFlags
10h agofreetimeout15002 ms————no_responsefetch_error:timeout
17h agofree402555 msv2yes$0.250x5E70…4EB9—
23h agofree402104 msv2yes$0.250x5E70…4EB9—
35h agofreenetwork333 ms————no_responsefetch_error:network
41h agofree402101 msv2yes$0.250x5E70…4EB9—
2d agofree402120 msv2yes$0.250x5E70…4EB9—