菜单
Cluster single-cell RNA-seq data and annotate cell types. Covers normalization, HVG selection, PCA, neighbors, Leiden clustering, and marker-gene-based annotation. Use when the user has a cell-by-gene matrix and wants cell-type clusters.
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
Install, deploy, find, organize, audit, and clean up agent skills with the `skl` (skillshelf) CLI — one canonical library, deployed on demand. Use when the user says "install this skill", "add a skill from GitHub", "deploy/activate these skills", "what skills do I have", "where is this skill deployed", "my skills are scattered", "tag/organize my skills", "remove a skill", "is this skill out of date", or mentions skillshelf / `skl`. Also when they say "skill" loosely about managing reusable agent instructions (not writing app code).
A demo multi-persona discussion skill. Three role-played experts — an optimist, a skeptic, and a moderator — debate a question and converge on a recommendation. Triggers: /sample-chatroom, "panel discussion", "debate this".
A demo authoring skill that produces a structured how-to guide with sections, callouts, and reference files. Use as an example of a multi-file skill with a README and a references/ directory. Triggers: /sample-guide, "write a guide", "how-to".
Deconstruct a vague term to its concrete usage. Use when a business or philosophical term is fuzzy and needs to be pinned down before reasoning with it.
Map collected artifacts to a checklist of required items and flag what is missing. Use when assembling or auditing an evidence package against a fixed set of requirements.
Pick a strong blog or social-post headline from a library of proven formulas. Given a topic, recommend the right formula, fill it, and explain why it fits. Use when writing or optimizing a post headline.
| name | scrna-cluster |
| description | Cluster single-cell RNA-seq data and annotate cell types. Covers normalization, HVG selection, PCA, neighbors, Leiden clustering, and marker-gene-based annotation. Use when the user has a cell-by-gene matrix and wants cell-type clusters. |
| domains | ["bioinfo","scrna"] |
Standard Scanpy clustering + annotation pipeline.
