the problem a long migration doesn’t fit in one agent chat. the context fills up and you start over. the fix cursor launched projects in beta on september 10. you talk to a coordinator agent that plans the work and hands it to other agents. the coordinator doesn’t write code itself. the project runs on its own cloud computer, so closing your laptop doesn’t stop it, and it keeps shared context files every agent can read. action open the agents window, pick projects in the sidebar and click new project. name it after the outcome, like “checkout redesign”, not the repo. pick the repo and the coordinator’s model. make your first message research only: ask for a plan, a conventions file and a how-to-test file before any code. tradeoff pro: when one agent works out how to test a service, later agents reuse those notes. con: cursor’s docs say five parallel subagents use about five times the tokens of one, and simple tasks can run slower. the “30% more prs merged” figure is cursor’s own. takeaway use it for work that takes weeks. a one-hour bug fix doesn’t need a coordinator.

the problem kept shipping agent output that looked right and wasn't. no way to tell the difference until a human read every line. the fix anthropic's outcomes feature, announced at code with claude sf in may 2026, adds a separate grading agent. you write a rubric describing what "good" means for the task, and a second instance of claude scores each run against it before anything gets delivered. action write the rubric before you build the agent, not after. list 3-5 concrete pass/fail criteria for the output, then let the grader enforce them on every run. tradeoff pro: turns "does this feel right" into a score you can track and improve against. con: a vague rubric gives you a confident, meaningless score, the grader is only as good as the criteria you wrote. takeaway stopped eyeballing agent output. wrote five criteria, the grader caught what i was skimming past.

people keep asking what's actually in my ai stack, so here's the honest breakdown. claude's about half my usage, writing, editing, thinking things through out loud. chatgpt's maybe a third, mostly for doing rather than thinking, quick research and image stuff. perplexity when i actually need a citation instead of an opinion. gemini only when i'm already stuck inside google docs or gmail, otherwise i forget it exists. but the split isn't the part i'd actually recommend copying. it's what i don't do: no single tool for everything, no first draft goes out unedited, nothing gets published without reading it out loud first. the percentages will look different in six months. that last part probably won't.