Included with Shogo Basic and upIntermediate · 8 lessons · 53 min
Agent Team Playbook
Hand your dev work to a team of agents, and stay the one who decides.
Eight guided steps from how Russell runs Shogo’s own engineering: a planner, three developers and a QA agent working in one channel, then the habits that keep them honest: cheap models for cheap work, self-checks, hard review questions, lessons, coaching, cleanup and evals. Each step ends with a setup prompt you paste into Shogo.
Lessons
- 1Run your dev work with a team of agentsFree preview✓ DoneOne channel where you post ideas in plain words, and a team of agents (a planner, three developers and QA) that turns each one into a planned card, a tested branch and a pull request, handing work to each other in the open. You make…5 min
- 2Stop paying the big model to do the cheap work✓ DoneWhat you'll have: a team split by role across two models — the expensive one plans and reviews, the cheap one executes — and the one sentence you put in the plan so the cheap model doesn't start second-guessing…5 min
- 3Give your agent a way to check its own work✓ DoneAn agent that, before it says "done", opens your app, looks at it at phone width, hits the endpoint with `curl`, runs the test that proves the change, and shows you all three in its reply. You judge the change from a screenshot…8 min
- 4Make the hard questions an agent's job✓ DoneA dev team of agents where every card has to answer four questions before it reaches you (*Is this the safest way? Is it the cheapest way? What could this break? Which tests prove it works?*) and a one-block Judgement from QA that…7 min
- 5Make your agent write down what it got wrong✓ DoneWhat you'll have: a small file of one-line lessons that every task reads at startup — plus the three questions that stop a "lesson" from being a restatement of the bug you just…6 min
- 6The expensive model is the coach, not the typist✓ DoneAn agent called Reflection, powered by your biggest model (I use Opus 5.5), that, every morning, reads everything your agent team did in the last 24 hours, asks "how did this happen?" about every mistake, and writes a short lesson…6 min
- 7Agents ship it, another agent cleans it up✓ DoneA cleanup agent that goes back over every card your agent team finished, picks up the crumbs (dead code, a helper written twice, a component nobody renders, code that works but takes a minute to understand) and writes the tests that…5 min
- 8Turn a real moment from your workspace into an eval✓ DoneA small suite of evals built from moments that actually happened in your workspace. Each one is a prompt, a seeded workspace, mocked tools, criteria scored in three phases, and a list of anti-patterns. You re-run the suite whenever…11 min