Medical Qa
- 16 installs
- 869 repo stars
- Updated June 8, 2026
- beita6969/scienceclaw
medical-qa is a Claude skill for medical question answering using structured biomedical knowledge bases and clinical datasets.
About
This skill defines a protocol for answering medical and biomedical questions using structured datasets like MedQuAD and PubMedQA. It parses the question, identifies sources, retrieves evidence, synthesizes an answer with a confidence level, and verifies against at least two sources. It never provides personalized medical advice.
- Evidence-based medical question answering over biomedical knowledge bases
- Uses MedQuAD (47,457 QA pairs) and PubMedQA datasets
- 5-step protocol with confidence levels and mandatory primary-source citations
Medical Qa by the numbers
- 16 all-time installs (skills.sh)
- Ranked #1,318 of 2,065 Data Science & ML skills by installs in the Skillselion catalog
- Data as of Aug 2, 2026 (Skillselion catalog sync)
medical-qa capabilities & compatibility
Free; relies on public biomedical datasets.
- Capabilities
- meta analysis · materials project
- Use cases
- research · data analysis
- Pricing
- Free
What medical-qa says it does
Medical question answering using structured biomedical knowledge bases and clinical datasets
**MedQuAD** (abachaa/MedQuAD): 47,457 QA pairs from 12 NIH sources
Never provide personalized medical advice
npx skills add https://github.com/beita6969/scienceclaw --skill medical-qaAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 16 |
|---|---|
| repo stars | ★ 869 |
| Last updated | June 8, 2026 |
| Repository | beita6969/scienceclaw ↗ |
What it does
Answer medical and biomedical questions with cited evidence and confidence levels using structured clinical datasets.
Who is it for?
Answering disease, drug, genetic, and prevention questions with cited evidence.
Skip if: Personalized medical advice.
When should I use this skill?
A user asks a medical or biomedical question that needs evidence-based, cited answers.
What you get
Returns a cited, confidence-rated answer cross-referenced against at least two independent sources.
- Cited, confidence-rated medical answers
By the numbers
- MedQuAD contains 47,457 QA pairs from 12 NIH sources
- 5-step answering protocol
- 3 confidence levels (HIGH/MODERATE/LOW)
Files
Medical Question Answering
Purpose
Answer medical and biomedical questions with evidence-based precision using structured datasets and clinical knowledge bases.
Key Datasets
- MedQuAD (abachaa/MedQuAD): 47,457 QA pairs from 12 NIH sources (NCI, GARD, GHR, MedlinePlus, NIDDK, NHLBI, NICHD, NIA, NIAMS, NINDS, NIDA, GARD)
- PubMedQA (qiaojin/PubMedQA): Yes/No/Maybe reasoning from PubMed abstracts
Protocol
1. Parse the question — Identify medical entities (diseases, drugs, genes, symptoms) 2. Source identification — Match question type to appropriate NIH source 3. Evidence retrieval — Search PubMed, clinical guidelines, drug databases 4. Answer synthesis — Provide answer with confidence level and citations 5. Verification — Cross-reference with at least 2 independent sources
Question Types
- Disease/condition: Etiology, diagnosis, prognosis, treatment
- Drug/treatment: Mechanism, dosage, side effects, interactions
- Genetic: Gene function, variants, inheritance patterns
- Prevention: Risk factors, screening, lifestyle modifications
Rules
- Always cite primary sources (PMID, DOI, or guideline reference)
- Distinguish between established evidence and emerging research
- Flag when evidence is limited or conflicting
- Never provide personalized medical advice
- Include confidence level: HIGH (multiple RCTs), MODERATE (observational), LOW (case reports)
Related skills
Microsoft FoundryDeploy, evaluate, and continuously improve Microsoft Foundry agents from a single agent interface.490k1.3k
Ai Research ReproductionOrchestrate trustworthy, auditable reproduction of deep learning repositories directly from their READMEs.176k511Run TrainSafely execute selected deep learning training commands with standardized evidence capture.176k511Explore RunSafely run isolated exploratory experiments with clear recording and conservative selection before committing changes.176k511Paper Context ResolverFetch precise reproduction-critical details like dataset splits, preprocessing steps, or evaluation protocols from the original academic paper when the repo README leav141k511Repo Intake And PlanScan unfamiliar AI research repositories and receive a minimal, trustworthy reproduction target before investing significant time.140k511FAQ
Which datasets does it use?
MedQuAD with 47,457 QA pairs from 12 NIH sources and PubMedQA for Yes/No/Maybe reasoning.
Does it give personal medical advice?
No, it never provides personalized medical advice and always cites primary sources.