Showing AI agent work in UI requires progressive disclosure — four design patterns emerging¶
Insight: As AI products lean into agentic capabilities, four design patterns are emerging for displaying agent work: (1) grouped results with sidebar expansion, (2) progressive disclosure with timeline, (3) dual-column collapse pattern where the process column collapses to summary once complete, and (4) expandable single-line indicators. The fundamental tension: "some users find the agent's work overwhelming and want the interface to focus purely on results. Others say seeing that work is essential for monitoring."
Detail: Luke Wroblewski identifies the dual-column collapse pattern (used by ChatDB) as particularly effective — the left column shows agent decisions and tool selection during execution while results appear in the right column, then the left side collapses to a summary once complete. This "allows the agent's work to serve a detailed progress indicator, instead of forcing people to watch a spinner." Key design recommendations: reduce visual weight as tool volume increases, show less process by default, shift focus to output once results appear, and enable optional review for trust-building users without forcing it on all.
Sources
- LukeW — "Showing the Work of Agents in UI" (2026-02-17)
Related: designing-agents-ontological-redesign in ai-assisted-design.md — COMPLEMENTS