菜单
Use when coding interviews, focus groups, or open-ended responses with AI assistance, when a codebook needs building or auditing, when intercoder reliability must be computed, or when qualitative quotes need integrity checks before publication.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when a complete research workflow — its instructions, skills, data boundaries, and tool connections — must be assessed for portability across AI agents or environments, so each component is explicitly mapped before migration rather than assumed to transfer.
Use when a Claude Code session loops, drifts from scope, claims a fix that CI rejects, hits context, PATH, or permission walls, or fails in a way another retry will not solve and a root cause diagnosis is needed instead.
Use when AI contribution metadata must match the actual workflow, when model alias and dated identifiers need auditing, when verified and fabricated citation counts feed a release decision, or when a venue's AI disclosure statement needs evidence behind it.
Use when a draft in Turkish, English, or both reads as AI-generated and needs revision that removes machine writing patterns and translation calques while preserving meaning, citations, statistics, and the author's voice.
Use when a reference list needs APA 7 cleanup, when DOIs must be confirmed real against Crossref or PubMed, when the user asks whether AI-suggested citations actually exist, or when a manuscript needs a citation hygiene pass before submission or defense.
Use when authorship order, CRediT contribution roles, or AI-assistance attribution must be recorded transparently for a manuscript, when an author dispute needs an evidence trail, or before a contributorship statement is required at submission; this skill maintains the human-attribution ledger of who contributed what and in which role (authorship order, CRediT taxonomy, revision responsibilities), which is distinct from the disclosure-metadata audit owned by ai-disclosure-auditor.
| name | qualitative-coding-discipline |
| description | Use when coding interviews, focus groups, or open-ended responses with AI assistance, when a codebook needs building or auditing, when intercoder reliability must be computed, or when qualitative quotes need integrity checks before publication. |
Use this skill when qualitative material is being coded and an AI tool participates in any part of it. The discipline keeps interpretive authority with the researcher while letting the tool earn its place as a second coder. It also applies retroactively, when an already coded dataset needs an audit before the findings are written up. It is not for anonymizing raw transcripts before coding begins; that is sensitive-data-anonymization-gate, and it is not for registering the analysis plan before coding starts; that is preregistration-analysis-plan-ledger.
Return:
Identifiable transcripts never enter the session. The tool never serves as sole coder, and a finding supported only by tool-generated coding is not reportable. Where the data are clinical or otherwise sensitive, the ethics approval and the data protection assessment govern what may be processed, and this skill defers to them.
"On iki görüşmelik veri setimde tematik analiz yapıyorum, kod defterimi denetle ve yapay zekâyı ikinci kodlayıcı olarak nasıl kullanacağımı kur."
Expected smoke output:
Bu beceri, nitel kodlamada yorum yetkisini araştırmacıda tutar. Kod defterinin sahibi araştırmacıdır, araç ancak ikinci kodlayıcı olarak çalışır ve uyuşmazlıklar tartışılıp tanımlar yenilendikten sonra uyuşma yeniden hesaplanır. Kaynağı gösterilemeyen alıntı yok hükmündedir. Aracın yorumu değiştirdiği her nokta refleksivite günlüğüne işlenir. Ham dökümlerin anonimleştirilmesi sensitive-data-anonymization-gate becerisinin işidir. Anonimleştirme tamamlanmadan ve onam kapsamı doğrulanmadan hiçbir katılımcı metni oturuma alınamaz — bu adım takdir meselesi değil, iş akışının sert ön koşuludur.
