How we protect stability, compatibility, and historical decisions through clear language

_{—Part of CSS Primer for the AI Era — CSS Communication Toolkit}

AI is good at adding, improving, and expanding
— but no so good at knowing when to stop.

If we do not name what must remain unchanged,
AI assumes everything is open for improvement.

Things like:

• structure
• layout
• spacing
• naming
• hierarchy
• behavior
• patterns
• history

So, the next communication skill is simple:

We name
what must not change
so AI understands
the real boundaries of the system.

This is how we protect stability
while still moving forward.

Here's whow not to say it:

“Don’t break this”

“Keep this stable”

“Don’t change too much”

AI cannot interpret these phrases.
They describe fear, not constraints.

AI respects stability only when
we name the parts that must remain stable.

So, we name them clearly.
Let's look at some example.

Every system has four categories of non‑negotiables:

1. Structure
   The HTML, the layout, the architecture.
2. Behavior
   How the component responds, adapts, or interacts.
3. Visual Identity
   The design system’s tone, scale, and hierarchy.
4. Historical Decisions
   Legacy patterns that exist for reasons AI cannot see.

When we name these,
AI understands the real limits of the work.

Structure is the backbone of the system.

So, we say:

“Do not change the HTML.”

“Do not reorganize the layout.”

“Do not introduce new classes.”

“Do not alter the grid structure.”

“Do not change the component API.”

And we avoid saying:

“Be careful with the structure.”

“Try not to break the layout.”

Structure is not a suggestion.
It is a boundary.

Behavior is how the system lives.

We say:

“This must remain responsive at all breakpoints.”

“This element must stay in normal flow.”

“This component must not overlap adjacent content.”

“This interaction must remain unchanged.”

We avoid:

“Don’t break the mobile layout.”

“Keep this working.”

That's too abstract.

Behavior is not optional.
It is a requirement.

Visual identity is the coherence of the system.

We say:

“Use only the existing color scale.”

“Follow the spacing scale.”

“Do not change the typography.”

“Do not introduce new decorative elements.”

“Maintain the existing hierarchy.”

We avoid:

“Keep it consistent.”

“Don’t make it too different.”

That's too abstract.

Visual identity is not taste.
It is a system.

History is invisible to AI. So we must reveal it.

We say:

“This spacing scale is historical — do not modify it.”

“This naming pattern is legacy — keep it consistent.”

“This layout supports older content — do not refactor it.”

“This selector is fragile — avoid touching it.”

We avoid:

“This is old — be careful.”

History is not nostalgia.
It is compatibility.

A complete “must not change” statement
looks like this:

“We want this element centered.
Do not change the HTML.
Do not modify other spacing rules.
Keep the layout stable across all breakpoints.”

Or:

“We want this section quieter.
Use the existing color scale.
Do not introduce new shadows or borders.
Keep the typography unchanged.”

Or:

“We want a refinement to the rhythm.
The spacing scale is historical — do not alter it.
Stay within this component.
Do not unify patterns.”

This is how we protect the system
while still making progress.

We name what must not change
when:

• AI stops refactoring
• AI stops redesigning
• AI stops unifying patterns
• AI stops altering behavior
• AI stops touching fragile code
• AI stays within the real limits
• the system stays stable

Non‑negotiables are not restrictions.
They are the shape of the system.

When we describe what must not change:

• we protect the architecture
• we protect the design
• we protect the behavior
• we protect the history
• and we protect the collaboration

We name what must remain unchanged.
AI respects the boundary.
And the system stays whole.

• Suggested Next Reading: How to Communicate Style, Tone, and Restraint

_{Tony de Araujo —New York | Lisbon}

• Top↑
• Back to Topic ↔
• Home←