I tried so hard, and got so far, but in the end it doesn't even matter...
Suffice to say, when I read your comment I was a bit touched, almost to the verge of tears, because not only did you try to play this game, but you replayed it again and again in order to actually try to win. This deserves a proper response:
I was experimenting with the current status of the game, if it is possible to win or not? The goal was to reach 100 cells. I wrote a system to emulate player input to make replays easier (to enable inhuman reaction times even), and experimented with ideal player options.
The best setup was just barrier around the enemy cell and the poison, but even then the player cell did not go as far as needed to win.
Then I touched the parts that used random choice between free cells and possible parents, and implemented heuristics that favor certain behaviors, like let the cell that has more energy multiply (dies probably later), or choosing places that have more food or further to the original cell, or even walls (because then no division happens and the cell does not lose energy from it), but to no avail. The most alive cells in a single iteration was just above 90, and no matter how I tried to "cheat" the random (saying that it is possible if it was true random for events to happen in this particular setup that favors the player) the results stayed the same.
So in conclusion, the possibility to actually win in this build is close to 0, but I cannot prove or disprove the impossibility of it. That would need some major rewrite to search the graph of all possible game states to find a winning move (individually evaluating each random choice), which is not a job for GDScript. And even if I would found one winning state that is reachable in this initial setup, that would probably mean it has astronomically low chance to actually be played out by chance like that. Not even factoring in the behavior of the pseudo random number generator, that might make some patterns of actions impossible to reach, and an entirely different problem of finding a random seed that makes that desired outcome.
Maybe one day i explore the wast possibility space and come back with a result. Or just fix the game and make some actually playable levels.
We'll see after the voting ends...