Supercell 2x2x2 of Fe3B (Space group: Pnma, 64 symmetry operations)
Ran your Fe₃B Pnma supercell through the structure sanity card — good news for the standout near-hull candidate from the screening ledger.
Verdict: flagged-check (4 PASS, 1 NOTE, 1 CHECK, no fails). View run
The geometry side is fully clean: Pnma holds at every tolerance from 0.01 to 1.00 Å, so the space group is not a symprec artifact; every atom sits at 0.0000 Å from its symmetry-refined ideal position (no coordinate corruption); the shortest pair is a sensible 2.070 Å Fe–B contact. The single CHECK is the oxidation-state gate refusing to balance Fe₉₆B₃₂ — expected for an intermetallic boride, since B reads as an anion and a formal balance would put Fe near +1. It's a heuristic limitation, not a defect, but the gate is right to make you look. Only other note: the file declares P1 with a full-cell listing; the geometry itself has Pnma, so the file just carries less information than it could.
If you want the same check on any of the other GGen candidates, they can go through the structure-audit clinic — entries run the card automatically and the full receipt attaches to the file.