Open-Air Stability: I left a bar on the counter for about 12 days and nibbled on it periodically to see if it was still palatable. It dried out somewhat, but didn't reach a point where it might be considered inedible.
Packaged Stability: To my great disappointment some of the bars I kept wrapped in plastic began to mold after only a week. This is a major step backwards from my previous version, and entirely unacceptable based on my stated purpose of supplying a stable food for backpacking. My goal is at bare minimum a two week shelf life. (I can't see anyone packing more than two weeks of food at one time.) I believe the culprit is the addition of water. To make the current recipe more stable it should be baked longer. I suggest bumping the time up to 1.5 hours at 250F, or baking at 250F for one hour then dropping the temp to 100F for an additional hour. Alternatively, you could try mixing the recipe with less water, or no water, from the beginning. Of course you could always just forgo the plastic wrap too.
Hot Breakfast Application: I suggested to someone in the comments section that if the bars get inedibly dry and you have a camp stove on hand you could make a kind of porridge. I tested the application with reasonable success. With a little encouragement from a fork the crumbled bar dissolved nicely in hot water and thickened to a pleasant consistency. Surprisingly the mix was a little bland and could benefit from some salt and sugar.
Notes for the Next Iteration:
- Solve the molding problem by reducing or replacing water content. I may try using an oil instead. Coconut oil, for example, has been suggested to me as a nutrition-dense candidate.
- Solve the sweetness problem by reducing or replacing the sweetened condensed milk. Again an oil like coconut oil may suffice. Finding more savory ingredients like roasted nuts or dried fruits that aren't as sweet may help too.
- Experiment with peanut butter alternatives. Enough people asked about a peanut-free variant that I figured I should give it a try. I'm eying Nutella as a prime candidate. The thinner consistency of Nutella may also help produce a softer texture without needing to add water.
- Track and optimize a larger spread of nutrition values. Next time around I plan to track fiber, sodium, and some of the bigger vitamins and minerals. I'm also toying with the idea of developing a mathematical model to optimize the nutrition within volume constraints. As I mentioned in the comments it's hard to add or change ingredients without radically altering the per cup averages, so I want a better way to visualize that process.