38 Open-Source Skills Fix AI Healthcare Reasoning Gaps

AI agents frequently misapply healthcare and life sciences decision frameworks despite knowing the guidelines, producing dangerous silent failures in variant interpretation, claims processing, and imaging analysis. A new collection of 38 open-source agent skills across 11 HCLS domains addresses this by encoding structured domain decision procedures into markdown documents agents can use at inference time. The skills, released under MIT-0 license, work across 20+ platforms including Amazon Bedrock and Claude Code without customization. Agents equipped with these skills won 70–86 percent of head-to-head comparisons against unskilled agents, with the strongest gains in critical thinking tasks. Users can also customize and create their own skills for specific use cases.