Nanoresearch Ideation

You give it a research topic and it fans out across arXiv and Semantic Scholar with multiple query angles, pulls back relevant papers, then does gap analysis to suggest novel hypotheses. The output is a JSON file with everything from paper metadata to a ranked hypothesis with justification. It's useful when you're starting a research project and need to map the landscape quickly, though the hypothesis quality will depend heavily on how well you frame the initial topic. The multi-query approach is smart since academic search can be finicky about phrasing. Think of it as a research assistant that reads widely and connects dots, not a replacement for your own domain expertise.

Install to Claude Code

npx -y skills add openraiser/nanoresearch --skill nanoresearch-ideation --agent claude-code