We analyze ingredient logic, not just lists.
We verify slaughter requirements for meat.
We include dual evaluation modes (Strict vs Balanced).
We flag doubtful items instead of assuming they are safe.
We explain our reasoning clearly for every result.
