nx_hmaccounter_extvec_gate.nx

nx_hmaccounter_extvec_gate.nx -- HMAC-SHA-1 over an 8-byte COUNTER, vs RFC 4226 Appendix D Table 1. CLOSES THE LAST UNGRADED TABLE IN A DOCUMENT ALREADY ON DISK. RFC 4226 Appendix D publishes TWO tables; Table 2 (the HOTP codes) is graded by nx_hotp_extvec_gate, and Table 1 -- ten RAW HMAC-SHA-1 values keyed by counter -- sat unused. ★ACQUIRED IS NOT COVERED, and a document is not finished when one of its tables passes (third time this session: rfc6234/rfc9106 sat pinned-but-ungated, and RFC 4648 needed all four encodings before it was actually done). ★WHY THIS IS NOT REDUNDANT WITH RFC 2202. RFC 2202 exercises HMAC-SHA-1 over ASCII strings and repeated byte fills. Table 1 exercises it over an 8-BYTE BIG-ENDIAN COUNTER -- a message that is mostly ZERO bytes with a low-order value, and exactly the shape HOTP/TOTP feed it in production. * A PRIMITIVE VALIDATED ONLY ON THE INPUT SHAPES SOMEONE THOUGHT TO PUBLISH IS VALIDATED ONLY ON THOSE SHAPES. Zero-heavy inputs are where length-handling and padding bugs hide. The secret is read from the document (`Secret = 0x3132...3930`), never typed from recall. Document pinned to a CORROBORATED digest (sovereign fetch and .NET WebClient agree). license_tier: ORIGINAL expect_exit: 0