History is a property of the chain, not of when you started.

A continuous register is assumed to need a warm-up period: deploy a cron, wait a year, have a year of history. But CV-1 is a pure function of chain state at a height — so the opinion for every past height already exists and has merely never been evaluated. This register did not wait. It computed.

No exception across 46 sampled heights on Flare mainnet, and 1 height we could not establish

Every backing control held at the 45 heights where the evidence could be established, and one height produced a DISCLAIMER rather than a verdict — recorded as unknown, never rounded up to a pass. That is a finding about Flare mainnet and nothing more — the same procedure run backwards over Coston2 reports 42 exceptions, because the Core Vault's destination allowlist was empty for its first three months and the outflow control had nothing to check against.

Verify that separately, one call, no trust in us:

cast call --rpc-url https://coston2-api.flare.network/ext/C/rpc \
  --block 27444811 0x4CB40b0dBfbF239eC60C9bE1496A6c1aA29e429b \
  "getAllowedDestinationAddresses()(string[])"
[]

Fig. 1 — A clean series is not a stronger result than a dirty one. It is a different one, and it is only worth reading because the same controls demonstrably fire elsewhere.

Every sampled height, and what it said

Fig. 2 — XRPL public retention floors at ledger 32570 (2013-01-01); earlier periods are structurally unobtainable from any public server and are not claimed.

[ ← Back to CV-1, the live register ]

The skew bracket has never suppressed anything

Across all 46 sampled heights, zero candidate exceptions were suppressed by the bracket. That is either “no artifacts occurred” or “the bracket cannot suppress”, and the two are indistinguishable from the result alone — which is precisely the failure this project already shipped once, one level further down.

So the adjudicator is a pure exported function with its own tests, and it is checked against a disagreeing bracket, a wholly-disagreeing bracket, and an unreadable one. It suppresses in all three. Its zero-suppression record here is therefore a measured result about the data, not an untested assumption about the code. Maximum observed skew across the series was 6 seconds.