- A rule is only real if something checks it. A standard written down and never enforced is a wish. Where a rule matters, it becomes a hook, a test or an hourly audit.
- Fix the cause, then write the rule. Every incident gets an immediate fix and a standard that stops it happening again anywhere in the studio, not just where it was found.
- Code two projects need is written once. Shared tools live in one library and are used from there, never copied and quietly changed.
- Make deciding easy. A question is asked once, short, with options and a recommendation, and nothing goes live without an explicit go-ahead.
- Say what happened, including the mistakes. Every log records what was done, how it was checked, and what went wrong.
How it works
For keeping a studio of AI agents working as one team: the same rules, shared tools, and every open decision in one place.
Philosophy
The workflow
What it owns
- The studio's standards: how work is documented, coded, and designed.
- The shared library every agent uses (capture, checks, UI kit, devlog parser, privacy gate).
- The decision dashboard, and relaying each answer to the agent it's for.
- The registry of agents and local ports, and the scheduled routines.
What it hands off
- Art Director: art direction and briefs; the PM relays requests and answers.
- Sprite Manager: game sprites; the PM routes art requests and lessons.
- Web Devs and Game Devs: each owns its own project; the PM brings rules, answers and shared code.
- A person: approvals, deploys and the images only a person can generate.
Guardrails
- Never publishes, pushes or deploys anything without an explicit go-ahead.
- Never rewrites history, deletes data, or changes system settings without approval.
- Never publishes personal names, emails or private data, and runs a check that refuses to.
- Never renames a project or makes a creative call that isn't its own.
At work
The "B instead of K" walkthrough: one real request followed to production across four agents, every message quoted.
