Thanks for the fix. Perhaps we could take this opportunity to further improve the presentation here, and shift away from this ambiguous and unexplained notation. Perhaps a grammar fragment:

min-max-macro:

signed-type _MIN
signed-type _MAX
unsigned-type _MAX

signed-type:

sized-type-prefix size
one of INTMAX INTPTR PTRDIFF SIG_ATOMIC WCHAR WINT

unsigned-type:

U sized-type-prefix size
one of UINTMAX UINTPTR SIZE_MAX

sized-type-prefix: one of INT INT_FAST INT_LEAST

size: one of 8 16 32 64

I think just following the C standard may be better: define italic N
for {8 16 32 64} and just enumerate others literally. The list fits in
285 chars, while your grammer goes 299 chars. (These numbers exclude
markups, and all word separators (including indents) are counted as 1.
Expanding {8 16 32 64} goes 663 chars.)

INT_N__MIN INT_FAST_N__MIN INT_LEAST_N__MIN
INT_N__MAX INT_FAST_N__MAX INT_LEAST_N__MAX
UINT_N__MAX UINT_FAST_N__MAX UINT_LEAST_N__MAX
INTMAX_MIN INTPTR_MIN
INTMAX_MAX INTPTR_MAX
UINTMAX_MAX UINTPTR_MAX
PTRDIFF_MIN SIG_ATOMIC_MIN WCHAR_MIN WINT_MIN
PTRDIFF_MAX SIG_ATOMIC_MAX WCHAR_MAX WINT_MAX
SIZE_MAX

From C99 7.20.1 p2

In the following descriptions, the symbol N represents an unsigned
decimal integer with no leading zeros (e.g., 8 or 24, but not 04 or 048).

As a bonus, some macros will become literally searchable.