Research Issues

When you need to understand what's actually broken or missing in a corner of Japanese NLP, this skill cross-references the bundled awesome-japanese-nlp-resources dataset with live web searches to produce an issue report. It's built for gap analysis: it scores the local corpus for coverage, spots what's underrepresented, then queries arXiv and Japanese-language sources for recent challenge papers, error analyses, and open problems. The workflow is methodical, date-aware bash and python blocks that avoid hardcoding assumptions, and it handles both specific topics and general overviews if you pass no argument. Most useful when you're scoping research, deciding where to contribute, or need to cite current limitations in a proposal. It won't use the Read tool on the large JSON file, which is smart. The challenge-biased search phrases make it more focused than a generic lookup.

Install to Claude Code

npx -y skills add taishi-i/awesome-japanese-nlp-resources --skill research-issues --agent claude-code