Portal UX Agent -- A Plug-and-Play Engine for Rendering UIs from Natural Language Specifications

TL;DR

Portal UX Agent addresses the tension between expressive natural-language UI specifications and governance by introducing a schema-bounded, plug-and-play pipeline that separates high-level planning from deterministic rendering. The system formalizes intent as $c = f_{\theta}(s)$ and renders $y = g(c)$ from a vetted component inventory under a schema $\\Sigma$, ensuring auditability and safety. A mixed-methods evaluation framework combines automatic checks with an LLM-based rubric (intent alignment, semantic correctness, accessibility) and light human adjudication, tested across multi-domain portal scenarios. The results show reliable translation of intents into coherent, accessible UIs with strong compositionality, suggesting practical pathways to controllable UI generation in real-world design systems.

Abstract

The rapid appearance of large language models (LLMs) has led to systems that turn natural-language intent into real user interfaces (UIs). Free-form code generation maximizes expressiveness but often hurts reliability, security, and design-system compliance. In contrast, fully static UIs are easy to govern but lack adaptability. We present the Portal UX Agent, a practical middle way that makes bounded generation work: an LLM plans the UI at a high level, and a deterministic renderer assembles the final interface from a vetted set of components and layout templates. The agent maps intents to a typed composition-template and component specifications-constrained by a schema. This enables auditability, reuse, and safety while preserving flexibility. We also introduce a mixed-methods evaluation framework that combines automatic checks (coverage, property fidelity, layout, accessibility, performance) with an LLM-as-a-Judge rubric to assess semantic alignment and visual polish. Experiments on multi-domain portal scenarios show that the Portal UX Agent reliably turns intent into coherent, usable UIs and performs well on compositionality and clarity. This work advances agentic UI design by combining model-driven representations, plug-and-play rendering, and structured evaluation, paving the way for controllable and trustworthy UI generation.