Calculate the inverse of the matrix
\begin{bmatrix} -1& 1& ...& ...&1 \\ 1& -1& 1& ... &1 \\ ...& ...& ...& ...&1 \\ 1&1 &1 & ...&1 \\ 1& 1 &1 &1 &-1 \end{bmatrix}
$-1$ on the diagonal and $1$ on the rest.
The key I think is to perform a sequence of elementary transformations on
[ A | $I_{n}$ ] until we get [ $I_{n}$ | $A^{-1}$ ] but that seems to be complicated.
Let $e$ be the all one vector. We have
$$A=-2I+ee^T$$ By Sheman-Morrison formula: \begin{align}A^{-1}&=(-2I+ee^T)^{-1}\\&=-\frac12I-\frac{-\left(\frac12I\right)ee^T\left(-\frac12I\right)}{1+e^T\left( -\frac12I\right)e} \\ &=-\frac12I-\frac{\frac14ee^T}{1-\frac{n}2} \\ &=-\frac12I-\frac{ee^T}{4-2n} \end{align}
Hence, the off diagonal entries are $-\frac1{4-2n}$ and the diagonal entries are $-\frac12-\frac1{4-2n}$.
Remark: If $n=2$, the matrix is not invertible.