@2023Anitachecks…add-clinical-ai-agent-safety → master1 files · +1 −0updated 2w agoclinical-ai-agent-safety, a portable Claude Skill for clinical research and healthcare-adjacent AI agent workflows.Clinicians and medical AI builders increasingly use agentic tools for literature review, research note cleanup, documentation, and coding support. This skill gives those agents practical safety boundaries: do not invent evidence, keep research separate from medical advice, preserve uncertainty, avoid sensitive data, and recommend human clinical review when claims could affect health decisions.
Inspired by my open-source rulebook, Clinical AI Agent Skills: https://github.com/2023Anita/clinical-ai-agent-skills
Use clinical-ai-agent-safety. Review this draft paragraph and identify unsupported medical claims, missing citations, overconfident clinical recommendations, and any places where uncertainty should be preserved.
git diff --check./clinical-ai-agent-safety/Gentle follow-up on this small external skill listing. The validation and security checks are passing; happy to adjust the wording or placement if maintainers prefer a different category.