// EVALUATOR_KIT · NO_LOGIN_REQUIRED
// WHAT_IS_IN_IT
Limit
// WHO_RUNS_WHAT
| Test | What it establishes | Driven from |
|---|---|---|
| D1 | Multi-vendor failover drill | Portal — you |
| D2 | KeySync — transit KEK synced, signing keys honestly blocked | Portal — you |
| D3 | Sign after failover — primary path and audit trail intact | Portal — you |
| D4 | Real primary outage — routing fails over, keys honestly do not | VM — operator |
| D5 | Real secondary outage — production unaffected | VM — operator |
| D6 | Recovery both sides — secondary automatic, primary by restart | VM — operator |
| P1–P2 | Generate a classical anchor key, then a linked ML-DSA-65 key | Portal — you |
| P3–P5 | Post-quantum sign, verify, and classical sign on the same key family | Portal — you |
| P6 | Disallowed post-quantum modes correctly rejected, both denials at AUDIT level | Portal — you |
| P7 | Vendor path honesty — the dashboard states software, not hardware | Portal — you |
The resilience half passes only if both outage directions pass. One direction is not a bidirectional resilience result, and the acceptance plan records that outcome as incomplete rather than as passed with a caveat.
// WHAT_THE_KIT_DOES_NOT_PROVE
// WHAT_A_TOKEN_ADDS
Everything on this page describes an environment. A token is what lets you drive it: take a vendor away and watch routing fail over, or have a post-quantum request denied on policy and then allowed.
Everything above is checkable without contacting anyone, and is meant to be read first. The token exists because the remainder needs a provisioned environment — not because the evidence is being held back.