Scilab doesn't have a direct analogue of Python's fcn(*v) function call that would interpret the entries of v as multiple arguments.
If you want to be able to call your function as either fcn(1,2,3,4,5,6) or as v = 1:6; fcn(v), then you'll need to add this clause to its beginning:
function z=fcn(a,b,c,d,e,f)
if argn(2)==1 then
[a,b,c,d,e,f] = (a(1),a(2),a(3),a(4),a(5),a(6))
end
// rest of function
z = a+b+c+d+e+f
endfunction
Now v=1:6; fcn(v) returns 21, just like fcn(1,2,3,4,5,6) does.
The condition argn(2)==1 checks if the function received one parameter instead of expected 6; if this is the case, it assumes it to be a vector. If that vector doesn't have enough elements (6) for the tuple assignment, an error is thrown.