Similar Resources

When you're working with Japanese NLP and find a good tokenizer or sentiment library, you inevitably want to know what else is out there. This skill takes a repo name, URL, or tool like "mecab" or "fugashi" and hunts down alternatives. It starts by scoring everything in the bundled awesome-japanese-nlp-resources dataset using IDF-weighted semantic labels and category matching, then expands the search to the web, merges results, and re-ranks. The similarity engine is smart enough to handle partial names and different owners. Use it when you want alternatives or related tools for a specific repo, not when you're doing a broad topic search. The matching logic is surprisingly thorough, down to tokenizing descriptions and penalizing generic terms.

Install to Claude Code

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