An Empirical Study of Agent Skills for Healthcare: Practice, Gaps, and Governance

TL;DR

This paper empirically analyzes healthcare agent skills, revealing their focus on patient workflow automation, uneven coverage of clinical tasks, and limitations of technical risk measures in healthcare settings.

Contribution

It provides the first empirical analysis of healthcare agent skills, highlighting gaps and challenges in their deployment and safety considerations.

Findings

Public healthcare skills focus on workflow automation and monitoring.

Coverage of clinical tasks and lifecycle stages is uneven.

Technical risk measures do not reliably reflect clinical risks.

Abstract

Healthcare automation is shaped by local procedures and organizational constraints, so agent capabilities rarely transfer unchanged across settings. Agent skills, self-contained directories that package reusable procedures for AI agents, are emerging as a procedural layer for adapting healthcare agents across diverse healthcare settings. We present the first empirical analysis of healthcare agent skills, drawing on 557 healthcare-related skills filtered from 58,159 public skills on ClawHub and annotated along ten dimensions covering function, deployment context, autonomy, and safety. We find that public healthcare skills emphasize patient-facing workflow automation and monitoring rather than the diagnostic and treatment-oriented tasks foregrounded in healthcare-agent research; coverage of the healthcare lifecycle and specialized clinical inputs remains uneven; and general technical risk does not reliably capture clinical risk. These findings position healthcare skills as a procedural layer not yet addressed by current benchmarks and risk frameworks.