#frameworks
8 posts
C.R.A.P., translated for agents: two principles compile, two only approximate
A computability audit of Williams' four principles: contrast compiles to an exact check, repetition compiles once a token file exists, alignment degrades to a proxy, and proximity does not survive translation at all.
Can an AI Agent Verify Material Design 3 Compliance From Token Files Alone?
Material Design 3 publishes its design rules as open-source constants — state opacities, elevation levels, shape radii, and a color algorithm. A test of how much M3 an agent can verify as arithmetic, and where the 2026 Expressive shift ends the constant tables.
Can an AI Agent Perceive Atomic Design?
Which layers of Atomic Design can an AI agent detect from DOM signals, and where does machine perception end and human interpretation begin?
Gestalt Principles Compile to Geometry: What an Agent Can Actually Compute
A test of the 08-04 verdict that Gestalt 'prescribes nothing' for agents. Proximity, similarity, alignment, and common region compile to DOM arithmetic; figure-ground and closure do not.
Your Design System Is Unreadable to AI — Tokens Fix That
Design tokens encode design decisions as structured, machine-readable data — making token coverage a computable proxy for how readable your design system is to an AI agent.
The Two Design Laws That Compile to Arithmetic
Fitts's Law and Hick's Law encode into exact arithmetic an agent can compute from the DOM: one from geometry, one from a count of choices. A test of both.
Which Design Frameworks Survive Contact with an Agent?
A review of design frameworks graded on one axis: can an AI agent encode them as deterministic checks? Nielsen's heuristics are the case study; most of the canon fails the translation.
The Verifiable Grid: Why 8-Point Spacing Is the First Design Decision an AI Agent Can Get Right
Of all design decisions, spacing is the only one an AI agent can verify with absolute certainty. Here's why the 8-point grid is the foundation of agent-readable design.