aipoch/medical-research-skills

What it solves

This project provides a comprehensive library of over 550 specialized skills for AI agents to assist medical and biomedical researchers. It addresses the gap between general-purpose LLMs and the rigorous, domain-specific requirements of medical research, covering the entire workflow from literature discovery to manuscript publication.

How it works

The library consists of SKILL.md compatible files designed for use with agent platforms like Claude Code and Codex. These skills are reviewed and evaluated using a domain-specific audit framework called MedSkillAudit to ensure they meet medical research standards. The library is organized into core research areas:

• Evidence Insights: Tools for search strategy design, literature synthesis, and gap identification.
• Protocol Design: Tools for hypothesis generation, sample size calculation, and study planning.
• Data Analysis: Bioinformatics pipelines (R/Python), statistical modeling, and visualization.
• Academic Writing: Assistance with SCI manuscript drafting, reporting guideline compliance, and peer review responses.
• Drug Discovery: Workflows for molecular docking, cheminformatics, and systems pharmacology.

Who it’s for

Medical and biomedical researchers who use AI agents to automate or enhance their scientific workflow, as well as developers building medical research assistants.

Highlights

• Extensive Scope: 550+ skills covering literature discovery, omics analysis, clinical research, and drug discovery.
• MedSkillAudit: A dedicated audit framework used to validate the research skills.
• Standardized Compatibility: Designed for SKILL.md compatible agent platforms.
• Comprehensive Writing Support: Specific tools for CONSORT, STROBE, and PRISMA compliance and NIH grant drafting.