Your effective limits, each with the deciding layer

Your effective limits, every knob in one read - the numbers the teaching errors quote, available before the wall instead of at it. Each row carries the effective value for you in this org, the layer that decided it ("default", or an "org override"/"customer override" set by support - overrides can lower a limit as well as raise it), whether support can set it per-org today (settable: true), who can raise it (settable_by: "self_serve" = an org admin, at once, up to raisable_to; "support" = request_limit_increase files the ask and an operator decides; "platform" = deploy configuration), and one line on what it bounds. A row's hard_max, when present, is the platform hard cap no override can exceed (fleet hardware, not policy). A row's raisable_to, when present, is the most an org admin gets at once with request_limit_increase. Without it, or above it, that tool files a request and an operator decides. Rows with layer "platform" are platform-set job, build, and checkpoint knobs: deploy configuration, not per-org levers today.