{
 "_doc": "The lab's current claim and limits for one result, copied word for word from the portfolio's current-claims file at the line named here. A value this site does not print in full is replaced by its sha256 and byte count, with the reason; the excerpts beside it are copied word for word from the same value. Nothing here is written by hand.",
 "result": "wireless-pqc:unseeded-hunt-11200",
 "title": "An unseeded adversary searched 11,200 compound attacks and found zero escapes; an exhaustive enumeration of all 53,485 compounds of up to three attack families found none either",
 "claim": "An unseeded adversary searched 11,200 compound attacks over the whole design axis and found zero escapes, with every elite non-vacuous; an exhaustive enumeration of the same grammar, every compound of up to three attack families on all ten designs, 53,485 in all, also found zero escapes.",
 "limits": "Exhaustive only up to order three; budgeted above it. The enumeration covers every compound of one, two or three attack families (317, 4,870 and 48,298 compounds; 53,485 in all) drawn from each design's own composable alphabet, 31 to 33 of the 34 families of attacks.ATTACK_FAMILIES, on all ten designs, judged by the same frozen verifier as the hunt (discover._all_invariants_hold through qd_discover._Evaluator). A compound is a set of families applied simultaneously, their adversaries merged field by field by discover._merge_adversaries, so orderings and repeated families are not in the grammar (the hunt's genomes are sets too), and neither are attacks modelled outside that table: discover.py names roam_ticket_replay, puzzle_replay, offline_dictionary and online_guess as non-composable. Designs, attacks and verifier are the lab's own Python model, not hostapd or wpa_supplicant. The same enumeration with KRACK re-opened (idempotent_install off) finds 10 escapes among the 317 one-family compounds, and at every order some compounds break the naive stack, so the closure is not over no-ops. Compounds of four to ten families are covered only by the MAP-Elites hunt: 11,200 evaluations is a budget; zero escapes within it is a null result, not a theorem. A verifier crash counts as FAILURE, never as an escape. The earlier \"2,146,200 evaluations\" headline is no longer stated (Appendix A, D-5); the runnable floor is 5,600 and this run is 11,200.",
 "claim_excerpts": [],
 "limits_excerpts": [],
 "lab_commit": "c5f34eda4cc9f7f86308a5db1cfa75efcd4c6841",
 "source": {
  "file": "portfolio-control registry/CLAIMS_CURRENT.jsonl",
  "line": 62,
  "sha256": "81776df15a1498abec7228ea179c952d01c54e031ffd4144498a30660963074b"
 },
 "evidence": "/receipts/wireless-pqc/results/unseeded-hunt-11200.json",
 "copied_utc": "2026-10-03T16:11:27Z"
}