The dimmer

Has this loop earned the dark? Four switches, one verdict. The instrument from chapter 8, standalone for workshop use.

A loop earns fully automated status only if its check is cheap, runs at high frequency, and cannot be faked outprimary — the dimmer splits that into the four properties you can actually audit. The setting is held in the page URL as you toggle, so a configured dimmer can be shared as a link: set the switches for the loop under discussion, copy the address, and send the argument rather than a screenshot.

Instrument · has this loop earned the dark?

Test a single loop, never an organisation. Set each switch to what the loop's check actually has today, not what the roadmap promises it.

3/4 conditions met

All four: candidate for darkness.

The loop has earned an unattended trial. Start narrow, log every gate, and keep an automatic park condition.

Three, and the missing one is unfakeable: keep a human gate.

The check can still be gamed. Automate execution, but keep approval lit until the oracle is genuinely independent.

Anything less: keep it lit.

The verification system cannot yet absorb full autonomy. Narrow the loop or improve the oracle before increasing speed.

Back-pressure rule: you cannot automate judgment you cannot observe.

Two things the instrument deliberately does not do. It does not average: three of four is not “mostly dark”, it is a specific verdict about which gate stays lit. And it does not remember your organisation’s roadmap: the switches describe the check the loop has today. A property the roadmap promises is a property the loop does not have.

The reading behind each verdict is in chapter 8; what the darkening of a real pipeline measured is in the case studies.