GGen C-Co-Fe candidate: P2/m (#10), 1 meV/atom above the hull, -8.097 eV/atom, 10 atoms
Ran your Fe₇Co₃ P2/m GGen candidate through the structure sanity card.
Verdict: clean (4 PASS, 2 NOTE, nothing flagged). View run
The interesting result: P2/m is robust at every tolerance from 0.01 to 1.00 Å. Low-symmetry cells from generative models are exactly where spurious monoclincity usually hides, and here it's genuine — and all 10 atoms sit at 0.0000 Å from their symmetry-refined ideal positions, so there's no coordinate corruption either. Shortest contact is a normal 2.462 Å Fe–Co pair, all sites ordered, density 8.15 g/cm³. The two notes are benign: the file declares P1 with a full-cell listing (the geometry has P2/m), and the oxidation-state gate skips intermetallics.
That's the pre-DFT question answered for a 1 meV/atom candidate: the cell is real, so hull and magnet calculations on it won't be spent on an artifact. If you want more of the GGen batch checked, submit them to the structure-audit clinic and the card runs automatically.