It's a little divider to keep your pens and pencils from shifting around inside your form holder.

Here, we machine one from aluminum bar stock to hold 7 instruments, plus one extra in the gap between the holder and the edge of the case, which is enough for everything I currently keep in my forms holder with a bit of room for growth.

It is a pretty straightforwards machining project, other than one "trick" to get the bottom strap so thin; to make a copy, you'll need (at least) a 1/4" roundnose endmill and a 1/4" (square) endmill, plus a flat file and a small round file.

The concept, of course, does not require that it be machined... for a while, I used one made from heat-bent plastic (this is a replacement because that one was ugly), or it could be made from bent sheet metal or rapid prototyped.

As an experiment, I actually made four of these. Of course, I only need one, so the other three are for sale on Ebay Here, here, and here. (Unless they sell for a lot, I won't be making any more.)

Step 1: Design and Initial Machining

Design:

The largest diameter of any of my writing instruments was 0.46", so a slot width of 0.5" seemed reasonable.
Spacing the pens - and thus the slots - 0.75" center-to-center made it reasonably easy to get my fingers around them for insertion/removal.
A slot depth of 0.46"/2 + 0.04" = 0.27" seemed reasonable.
When thinning the bottom out, I successfully milled it down to 0.034" on one of the samples. I didn't try to go thinner because I was getting nervous, and I didn't do experiments to see just how far I could push this process. (Thinner would make the pens easier to insert/remove.)
The inside radius, 1/4", was chosen because that's the only size round nose endmill I have.
All lengthwise dimensions are multiples of 1/16" because my X/Y table moves that far per turn of the dial.

The initial (straightforwards) machining steps are as follows:

Cut a piece of 1/2"x1/2" aluminum bar to length: 5.5" plus some slop (say, 5.65").
Set up a vice parallel to the axis of your mill.
Using an endmill larger than 1/2", clean up the four long faces of the stock. (I used a 9/16" 4-flute)
You'll need to deburr the corners between each cut.
Clean up the left end of the stock.
Clean up the right end of the stock, and bring the total length to 5.50" (perhaps ± 0.01")
While not strictly required, you may want to layout the slots you will be cutting.
Remove most of the material in the slots with a 7/16" endmill centered on each slot at a depth of 0.2".
Machine the sides of the slots and the corner radii with a 1/4" round nose endmill. Depth of cut should be 0.27" below the tops of the tabs.
Machine the flat bottom of the slots using a 1/4" endmill. Depth of cut should be the same as for the round nose endmill, +0.000 / -0.002".

Notes on machining:

When making more than one of these, I found that it was faster to swap a part out of the vice and rezero the X dimension than it was to swap the cutter out and rezero the Y dimension, so doing this saves time if you can make each cut with one pass.
Hence why we "hog out" most of the material in the slots with a 7/16" endmill in one pass: this makes it so that we can do all the rest of the cuts in one pass too.
A 7/16" endmill at a depth of 0.2" only has a couple hundredths of space between it and the final profile. Be careful to locate it in the center of the slots.
My mill thinks that a 7/16" x 0.2" slot in aluminum is pretty big, and so prefers a 2-flute cutter for that operation.
I got the best surface finish on the corner radii by climb milling, which surprised me, given how lightweight my mill is.
It looks and feels way better if the flat bottom of the slots is a bit high than a bit low, hence the asymmetric tolerance on that feature.

Step 2: Finishing

Machining processes tend to leave burrs, and that's no fun, so you should remove them ("break all edges and corners").

You'll note that one of the pictures above shows the use of files after all machining processes are done - including thinning the bottom. This works, but is stupid and slow:

Deburr the edges of the slots before you thin the bottom, while the part is still strong and you can stick it in a vice.
For deburring the curved edge at the front and back of the slots, compared to a small round file, a 90° rotary file (at maybe 800 rpm) produces superior results in much less time. By clamping a drill in a vice (use a miniature bar clamp or small C-clamp to engage the trigger), I was able to set up so that I could sit with the work at eye level and my arms supported on a table, which was nearly ideal.

Step 3: Funny Machining and Installation.

Once the slots are deburred, there's not much to it:

Put the part bottom-side-up in the vice, making sure to press it down so that all of the tabs make contact with the parallels; clamp moderately so as not to dent the small tabs that you are contacting.
Working only on the supported side, make light passes until the you have the desired thickness. I started doing 0.03"/pass with moderate feed/tooth, then removed the final 0.045" with a 0.02", 0.015", and finally 0.01" depth of cut.
If your vice was short, and you couldn't support the whole workpiece in it at once, turn it around and do the other side by repeating steps #1 and #2. Make sure that the depth of the final passes matches up. (Call it a flatness tolerance of 0.003" with a clamping constraint, bottom surface of part to flat plate, of not more than 5 lbf/0.75" length... if you want to be ridiculously technical. :-p)

A couple notes:

I used 6061 T6511 aluminum. T6511 allows for straigtening post heat-treatment, which sets up internal stresses which, relieved when you remove half the bar, will cause the stock to warp slightly. Hence the note about making sure all the tabs are contacting the parallels, and one of the reasons we couldn't use an unconstrained flatness tolerance. (On the upside, T6511 is much cheaper that T651)
The pictures show a 2-flute endmill, but I also have used a 4-flute, and it worked fine. And removed material twice as fast for a given feet/tooth.

The completed part can be washed, dried, and installed in your form holder.
I might want to remove it for use elsewhere, so I used hot glue.
(AFAIK, the best way to hot glue small metal parts like this together is to apply a bit of glue to one, clamp them together using small clamps like binder clips, and then evenly heat the whole assembly with a propane torch or other heat source until the glue melts and the gap closes.)