I'm trying to find such primes that are of the form $x^6+y^6$; $x,y\in\mathbb{Z}$.
If one of $x$ and $y$ is $0$, say $y=0$ then $x^6+y^6=x^6$ which is not a prime. So assume that $(x,y)\ne(0,0).$
I factored $x^6+y^6$ as follows:
$x^6+y^6=(x^2)^3+(y^2)^3=(x^2+y^2)(x^4+y^4-x^2y^2)=(x^2+y^2)[(x^2+y^2)^2-3x^2y^2]=(x^2+y^2)(x^2+y^2+\sqrt{3}x^2y^2)(x^2+y^2-\sqrt{3}x^2y^2)$
If $x^4+y^4-x^2y^2=1$ then $(x^2+y^2+\sqrt{3}x^2y^2)(x^2+y^2-\sqrt{3}x^2y^2)=1$ this implies that the two numbers $x^2+y^2+\sqrt{3}x^2y^2$ and $x^2+y^2-\sqrt{3}x^2y^2$ are reciprocal to each other. So if $x^2+y^2+\sqrt{3}x^2y^2\in\mathbb{Z}$ then $x^2+y^2-\sqrt{3}x^2y^2\not\in\mathbb{Z}$ or vise-versa.
But $x^2+y^2\pm\sqrt{3}x^2y^2\not\in\mathbb{Z}$ for $(x,y)\ne(0,0)$.
So that $x^4+y^4-x^2y^2$ can't be factored out in $\mathbb{Z}$. That is we can conclude that this is a prime.
To $x^6+y^6$ be a prime, we must have $x^2+y^2=1$, which is possible only for $(x,y)=(+1,0);(-1,0);(0,+1);(0,-1)$.
But in all the above cases, $x^6+y^6=1$, which is not a prime.
So I can conclude that there is no prime of the form $x^6+y^6.$
My question is my argument ok? And is there any other way to think about this problem?
So you know that $x^6 + y^6 = (x^2 + y^2)(x^4 - x^2y^2 + y^4)$, and both are not factorizable over $\mathbb Q$.
If $a, b, c \in \mathbb Z$ and $a = bc$, then if $a$ is prime, either $b = 1$ or $c = 1$.
You already know that $x^2 + y^2 = 1$ if and only if $(x, y) \in \{(0, \pm 1), (\pm 1, 0)\}$. Let us briefly look at the other term.
Actually, $x^4 - x^2y^2 + y^4$ will not be small too often, since $x^4 - x^2y^2 + y^4 = (x^2-y^2)^2 + x^2y^2 \geq x^2y^2$. In particular, it is $1$ only if $(x, y) = (\pm1, \pm 1)$.
Combining all these, the only prime $x^6+y^6$ hits is $2$.
For your other question in your comment, actually one can prove more:
If you care about why, here is a sketch of the construction. The construction can roughly be taken in the following steps:
The iteration: For $i = 2, \cdots n$, determine the final value $a_i$ satisfying the following three conditions:
(a) $a_i \equiv 1 \pmod p$.
(b) $a_i$ is relatively prime to $a_1, \cdots, a_{i-1}$.
(c) $|f(a_1, \cdots, a_i, 1, \cdots, 1)| > p$
This is always possible. For the correctness, (a) guarantees us that $p \ | \ |f(a_1, \cdots, a_i, 1, \cdots, 1)|$; (b) guarantees everything are still relatively prime, and (c) guarantees that it is composite.
After the iteration we should get $(a_1, \cdots, a_n)$ satisfy the condition we requested.