MedSci Skills

Install

npx medsci-skills install

This is third-party code. Review the repository files before installing.

What it does

The collection takes a medical study from topic and literature search through design, sample size, data cleaning and statistics to figures, manuscript text, reviewer responses and submission prep. Its core is verification: check-reporting audits a manuscript item by item against guidelines such as STROBE, CONSORT, PRISMA and TRIPOD plus risk of bias tools, other skills catch fabricated or mismatched citations via PubMed and CrossRef and check numerical consistency. Version 5.0 adds a medical ML model lane: architecture choice, a reproducible PyTorch repo scaffold and validation audits. The orchestrate skill classifies a request and routes it to the right skill.

Who it is for. For physician-researchers and medical engineering investigators preparing papers for publication.

Example request

Check my cohort study manuscript against STROBE and flag missing items

Limitations

The author states plainly that it is not a diagnostic tool or an autonomous author: every output needs expert verification. Materials and reports are in English, though you can talk to the skills in your own language. Some checklists are educational summaries due to licensing. Public APIs such as PubMed and CrossRef need no keys.

How to disable. Disable the medsci-* plugins via /plugin or delete the skill folders from ~/.claude/skills.

Security check

• Runs analysis scripts and reads project files
• Calls public APIs such as PubMed and CrossRef
• May handle clinical data, with a dedicated de-identification skill

README in short

The README describes the collection's purpose, its scope boundary and the path from research topic to presentation. It offers several install paths: npx medsci-skills, copying from git, gh skill, a Claude Code marketplace with nine category plugins, and a ZIP installer for people who do not use a terminal. It mentions two skills published as standalone repos and three starter workflows. MIT licensed, with a DOI and a preprint.

SKILL.md

---
name: check-reporting
description: Check manuscript compliance with medical research reporting guidelines. Supports 49 guidelines including STROBE, STROBE-MR, RECORD, REMARK, TARGET, GATHER, CONSORT, CONSORT-AI, STARD, STARD-AI, TRIPOD, TRIPOD+AI, TRIPOD-LLM, PGS-RS, ARRIVE, PRISMA and more, plus risk of bias tools (QUADAS-3, QUADAS-2, RoB 2, ROBINS-I, PROBAST, NOS, COSMIN and others). Generates item-by-item assessment with PRESENT/MISSING/PARTIAL status.
tools: Read, Write, Edit, Bash, Grep, Glob
model: inherit
---

# Check-Reporting Skill

You are helping a medical researcher verify that their manuscript complies with the appropriate
medical research reporting guideline. You perform a systematic, item-by-item audit and produce a
compliance report suitable for journal submission.

## Communication Rules

- Communicate with the user in their preferred language.
- Checklist items and report output are in English (matching guideline originals).
- Medical terminology is always in English.

## Reference Files

- **Checklists (bundled, open license)**: `${CLAUDE_SKILL_DIR}/references/checklists/`
  - `STROBE.md` -- observational studies (CC BY)
  - `STROBE_MR.md` -- Mendelian randomization studies, STROBE-MR 2021
  - `STARD.md` -- diagnostic accuracy studies (CC BY 4.0)
  - `STARD_AI.md` -- AI diagnostic accuracy studies
  - `TRIPOD.md` -- prediction models, classic 2015 version
  - `TRIPOD_AI.md` -- prediction models with AI/ML (CC BY 4.0)
  - `TRIPOD_LLM.md` -- studies using large language models, TRIPOD-LLM 2025

Related