research-superpower
Systematic literature search, screening, and citation traversal
All Skills in This Repository (10)
Pure Emerald Level IndicatorsGetting Started with Research Superpowers
Automate systematic literature searching and review workflows for PubMed and Semantic Scholar.
Cleaning Up Research Sessions
Remove intermediate files from completed research sessions while preserving core outputs.
Building Paper Screening Rubrics
Develop and refine paper screening rubrics with versioned JSON criteria.
Answering Research Questions
Automate literature reviews with search, evaluation, and synthesis phases.
Traversing Citation Networks
Traverse citation networks via Semantic Scholar with depth-limited BFS and relevance filtering.
Evaluating Paper Relevance
Screen research papers by scoring abstracts and extracting requested data.
Checking ChEMBL for Structured SAR Data
Query ChEMBL by DOI to retrieve curated bioactivity data counts and metadata.
Subagent-Driven Literature Review
Orchestrate subagents for parallel literature screening and deep-dive extraction.
Searching Scientific Literature
Search PubMed and extract paper metadata into structured JSON.
Finding Open Access Papers
Query the Unpaywall REST API by DOI to return the best open-access PDF URL.
Frequently Asked Questions
FAQPage SchemaHow to install research-superpower?▼
Run `npx skills add kthorn/research-superpower --all -g -y` in your terminal to install all literature review skills globally.
How to do a systematic literature review with AI?▼
Ask a research question in plain English and the skills search PubMed, score abstracts, deep-dive into relevant papers, and synthesize findings into a SUMMARY.md file.
Can it find free versions of paywalled papers?▼
Yes. It automatically checks the Unpaywall API for open-access versions whenever a paper hits a paywall.
Does it work for medicinal chemistry research?▼
Yes. It checks the ChEMBL database for curated bioactivity data like IC50 values on medicinal chemistry papers before parsing PDFs.
Can it handle large searches with 100+ papers?▼
Yes. It uses validated screening rubrics and parallel subagent processing to screen large result sets quickly with progress checkpoints.
Related Repositories in Education & Research
View All in Education & Research→learn-claude-code
Learn to build AI agent harnesses from scratch
ai-agents-for-beginners
Learn to build, deploy, and secure AI agents step by step
33-js-concepts
Master 33 essential JavaScript concepts with tested, verified guides