Ordered playbooks for real goals

There are 100+ endpoints here. The spec tells you WHAT exists; it cannot tell you what order to call things in, or which checks it would be reckless to skip. That knowledge otherwise lives in prose written for humans, so every agent re-derives it badly and differently — quoting without checking serviceability, paying without confirming a recipient, reporting success when verification said undetermined. Each recipe is an ordered sequence: which endpoint, what it answers, what to do with the answer, whether it is optional, and what skipping it actually costs. Every referenced endpoint is checked against the live spec on each request — a recipe naming an endpoint that no longer exists is worse than none, because it will be followed.