Skip to content

CAPABILITY 02 / 30

Adapt, don’t design

The rule that keeps AI output on brand: start from an approved template, fill the slots, escalate the rest.

IN PRODUCTION

This runs in client systems today. The sample below was regenerated for a fictional company.

Left to itself a model will invent a layout, and it will be a good-looking layout that is not yours. The fix is architectural. A skill names the template it must use, the slots it may fill, and what it has to escalate. The facts change. The frame does not.

TRY IT / RUNS IN YOUR BROWSER

Use it here.

The mechanism is the production one. The company, the catalogue, and every figure are fictional.

Loading the skill anatomy.

THE MECHANISM / SHOW THE SHAPE, NOT THE SOURCE

How it works.

What a skill is
  1. 01Choose

    The skill selects an approved template, it does not compose one.

  2. 02Fill

    Only defined slots take content, and each slot has a length and a type.

  3. 03Refuse

    Content that does not fit a slot is rejected rather than squeezed in.

  4. 04Escalate

    A genuinely new layout is a request to a person, not a decision for the model.

  5. 05Record

    The output records which template and which tokens produced it.

WHAT IT READS

  • An approved template library
  • A slot definition per template
  • A rule for what may never be invented

WHAT IT CHECKS BEFORE IT SHIPS

  • No new colour, font, or spacing value appears in an output
  • Heading order matches the template
  • A rejected fit is reported, not silently trimmed
  • Every output names the template it used
In production: the anatomy plus swap-mark pattern is the standard skill shape in both company systems.

The mechanism runs inside client systems. Everything on this page was regenerated for Apex Instruments, a company that does not exist. Client names, source files, and internal wording stay private.