Does Special Relativity require a "ruler postulate" analogous to the "clock postulate"? It's fairly well known that the clock postulate is needed in Special Relativity when dealing with accelerated clocks, so does something analogous exist when dealing with accelerated spatial displacements?
By this I mean: the spatial distance between two identically co-accelerated points doesn't change in the instantaneous co-moving inertial frame.
 A: I think you have slightly misinterpreted what the clock postulate says, and in doing so you have introduced an artificial distinction between time and length.
I think special relativity is best understood as a vacuum solution to Einstein's equation. Viewed this way, special relativity is a spacetime where the geometry is given by the Minkowski metric:
$$ ds^2 = -c^2dt^2 + dx^2 + dy^2 + dz^2 $$
The quantity $ds$ has units of length, and it's also often written as $ds^2 = -c^2d\tau^2$ where $d\tau$ has units of time and is called the proper time. The interpretation of $ds$ is that if you have some curve in spacetime then you can integrate $ds$ along the curve to give the length of the curve. For the observer moving along the curve, in their own locally inertial frame they are stationary so $dx = dy = dz = 0$ and so:
$$ ds^2 = -c^2d\tau^2 = -c^2dt^2 $$
and integrating this gives $s = ct$ or more intuitively $\tau = t$ i.e. the proper time, $\tau$, along the curve is the same as the time, $t$, shown on the clock carried by the moving observer.
The clock postulate says this is true whether the curve is a geodesic and the observer is moving freely, or whether the observer is being accelerated. Presumably it's called the clock postulate because it relates to the calculation of the proper time along the curve. But the quantity being calculated is the path length along the curve, and the proper time is simply another way of giving the length but divided by $c$. So the clock postulate could just as accurately be called the length postulate. Whatever you call it, it is simply the length of a curve.
This is why you do not need a separate ruler postulate, because the clock postulate is both a time and a length postulate depending on whether you're calculating $\tau$ or $s$.
