Behind the Craft · Sunday, September 27, 2026
Lauren explained how she uses Grockbot to manage her personal stack of plugins, P-Stack. She initiates a 'cursor Cloud agent' which performs automated evaluations ('Evals') to ensure skill improvements result in better performance. Only after the Evals playbook in P-Stack confirms the goal was achieved does the system finalize the pull request.
“Um, so, uh, I'll kick off Cloud agent, cursor Cloud agent, uh, which will, um, essentially do Evals for me.”
“So in order to, like make a change to my set of skills that I can trust that the skill improvement is actually going to result in better performance.”
“Uh, I have this Evals playbook in P stack. Uh, and then it will, you know, like, uh, try the prompt, uh, it will, you know, like, uh, make sure that the, the intent, the goal that we actually set out to achieve was achieved.”