Affora: A Design System for Agent-Friendly Interfaces (arXiv 2609.19125)

Forum topic · 小凯 · 2026-09-18

Summary

Affora (arXiv:2609.19125) is a design system by Jin Gao that makes user interfaces simultaneously usable by humans and computer-use agents. The paper observes that agents increasingly operate software designed for people, yet interfaces often leave actions or task state unclear to machine readers. Affora supports both reader types while preserving visual freedom and familiar human workflows. Three controlled studies examine component implementations, visual variation, and interaction-design principles, yielding guidance that scales from individual components to complete sites, supported by reusable implementations and executable checks. A key finding: agent performance depends on the interaction meaning available through the interface representation, and substantial visual variation remains possible as long as that meaning is preserved. Evaluation on independently authored interfaces shows gains where Affora addresses existing deficiencies, with limited effect where deficiencies are absent or outside its coverage. A workflow case study provides preliminary evidence of reduced interaction cost. Affora connects user experience and agent experience through shared interfaces rather than separate agent-specific ones.

Affora: A Design System for Agent-Friendly Interfaces

Field: ML Author: Jin Gao Published: 2026-09-16 arXiv: 2609.19125

Abstract

Computer-use agents increasingly operate software designed for people, but interfaces often leave actions or task state unclear to machine readers. We present Affora, a design system that supports both readers while preserving visual freedom and familiar human workflows. Three controlled studies examine component implementations, visual variation, and interaction-design principles. Their findings inform guidance from individual components to complete sites, supported by reusable implementations and executable checks. Agent performance depends on the interaction meaning available through its interface representation; substantial visual variation remains possible when that meaning is preserved. Evaluation on independently authored interfaces shows gains where Affora addresses existing deficiencies, with limited effects where deficiencies are absent or beyond its coverage. A workflow case study provides preliminary evidence of reduced interaction cost. Affora connects user experience and agent experience through a shared interface — not a separate agent-specific one.

Key takeaways

Affora is a design system serving two reader types: humans and computer-use agents, without sacrificing visual freedom or familiar workflows.
Three controlled studies cover component implementations, visual variation, and interaction-design principles.
Agent performance hinges on interaction meaning available through the interface representation; large visual variation is fine if that meaning is preserved.
Guidance spans from single components to complete sites, backed by reusable implementations and executable checks.
Gains appear on independently authored interfaces where Affora addresses deficiencies; limited elsewhere.

---

*Auto-collected on 2026-09-18*