Loops Housevet402: x402 on NEAR, counted against a pre-registered line

NEAR HACKATHON
Links
Team
1 member- TAOwner
Takeshi Kazumoto
Overview
x402scan can't count x402 payments on NEAR. So I posted a line on chain (25,400) before counting: at most 2,532.
This is a pre-registered investment thesis. N counts every delegate-action ft_transfer of NEAR USDC, x402 or not, so it is an upper bound on x402 exact-scheme payments. The line N ≥ 25,400 went on chain before the count. The result is N ≤ 2,532 < 25,400.
Known relay accounts sum to 1,716, a lower bound on N and 6.8% of the line. Named x402 facilitators settled 0 exact-scheme payments from 2026-10-08 11:42 to 2026-11-07 11:42 UTC.
I claim NEAR's edge in agent payments is the funding and settlement layer. NEAR Intents volume was $3.97B over 30 days (DefiLlama totalDataChart, 2026-09-29 to 2026-10-29 UTC). That volume is not a count of agent payments.
Every figure rebuilds from a frozen public snapshot, and a live dashboard shows the daily reading. The repository also ships a reference facilitator for the NEAR exact scheme with a delivery receipt.
To reproduce with no network, run node thesis.mjs --offline on Node 20 or newer, with nothing to install. The line it prints starts with "rebuilt 77 numbers from data/snapshot (no network), 2 checks ok" and includes "N ≤ 2,532 < 25,400: the threshold is not crossed".
Dashboard: https://kzmttkc.github.io/vet402-near-thesis/ Walkthrough video: https://youtu.be/FH0V72KDRiw
vet402 is a product I run, and a finding that delivery records matter would help it.