I don't think any flex hose manufacturer will bat an eye if we're talking about a few MM associated with installing an anchor at an almost immediate location from the flex hose, and would likely wholeheartedly prefer it as it eliminates as much piping load as possible onto their unit.
If we're talking about a typical steel nozzles on an API tank, you should have adequate load capacity with the addition of the flex hose. It's common for nozzles to not be analyzed with a properly installed flex hose (see manufacturer's recommendations).
I speculate the manufacturer's concern is that fully axially loading the hose (i.e. when the braids bend and lock against one another) will have deleterious effects on the pressure and lateral/angular displacement capacity of the hose.
Allowable loads for a given nozzle is an 8 dimensional problem (Fx, Fy, Fz, Mx, My, Mz, P, T), which sometimes can be brought down to a 6 dimensional problem (Fr, Fa, Mr, Ma, P, T) depending on symmetry, a flex hose is also the same kind of 6 dimensional problem, except where axial loading has a much faster deleterious effect on the others. (In this case, I treat forces and displacements as analogous, but also understand that they are not necessarily linear with each other.)