Claude Skills 中文技能市场——为 Claude Code、Codex 与 Cursor 精选 Agent Skills
收录 90,000+ 开源 SKILL.md 工作流,提供中文解读、安全检测与高速下载,一行命令即可装到你的 AI 编程助手里。
高分中文技能
osta-docs-local
OSTA 本地文档(OSTA/)
literature-search
全面文献检索,自动摘要,趋势分析和关键论文识别,应用于文献综述、趋势识别、关键论文发现以及跟踪研究进展。关键词:文献、PubMed、论文、搜索、文献综述、回顾
structure-search
基于结构的相似性搜索和骨架分析,用于药物发现。适用于先导化合物跳跃、骨架形态变化和化学空间探索。关键词:相似性搜索、骨架跳跃、化学空间、指纹、Tanimoto系数
packmol
This skill should be used when the user asks to "create a packmol input", "pack molecules with packmol", "solvate a protein", "build an initial configuration", "setup molecular dynamics", or discusses molecular packing, solvation, or building simulation starting structures.
hypogenic
使用大型语言模型自动生成假设并进行测试。在从数据集中生成科学假设、将文献见解与经验数据相结合、根据观察数据测试假设或在欺骗检测、人工智能内容检测、心理健康分析或其他实证研究任务等领域进行系统假设探索以发现研究发现时,可以使用此技能。
research-with-citations
Conduct thorough research with proper citations and source verification for academic and professional content
research-intelligence
Extract insights, analyze claims, and synthesize knowledge from research content. Use when processing academic papers, articles, podcasts, videos, meeting transcripts, or any content where the goal is to extract wisdom, analyze arguments, summarize findings, or compile references. Triggers include "analyze this paper", "extract key insights", "summarize the research", "what are the main claims", "extract wisdom from", "compile references", "critique this argument".
bayesian-meta-analysis
教授贝叶斯元分析方法,包括先验指定、MCMC方法和后验分布的解读。适用于用户需要整合先验知识、需要概率解释或处理稀疏数据时使用。
heterogeneity-analysis
Assess and interpret between-study heterogeneity in meta-analysis using I², Q statistic, tau², and prediction intervals. Use when users need to evaluate consistency across studies, understand sources of variation, or decide if pooling is appropriate.
forest-plot-creation
Generate and interpret forest plots for meta-analysis visualization using R and the metafor package. Use when users need to create forest plots, understand visual representation of pooled effects, or interpret study weights and confidence intervals.
grade-assessment
Apply the GRADE framework to assess certainty of evidence in systematic reviews. Use when users need to rate evidence quality, create Summary of Findings tables, or understand the factors that affect confidence in effect estimates.
diagnostic-meta-analysis
教授诊断测试准确性研究的元分析,包括灵敏度、特异性、SROC 曲线和双变量模型。当用户需要综合诊断准确性数据、理解 SROC 曲线或使用 QUADAS-2 评估质量时使用。
publication-bias-detection
使用漏斗图、Egger检验、剪枝填充法和选择模型检测和评估荟萃分析中的发表偏倚。当用户需要评估缺失研究是否会影响其结论时使用。
meta-analysis-fundamentals
Teach the foundational concepts of meta-analysis including effect sizes, statistical models, and evidence synthesis. Use when users ask about meta-analysis basics, want to understand pooled effects, or need guidance on fixed vs random effects models.
parallel-literature-search
Parallel search across PubMed, Perplexity, and your knowledge base. Searches all sources simultaneously and synthesizes findings with citations. Faster evidence gathering for clinical questions.
searching-academic-outputs-with-dimensions
Search for academic literature, empirical evidence, and scholarly research using the Dimensions database. Use when seeking research papers to support product decisions, find empirical studies, conduct literature reviews, explore funding patterns, validate hypotheses with academic sources, or discover research trends. Supports publications, grants, patents, clinical trials, and researcher profiles. Triggers on requests for "academic evidence", "empirical research", "find studies", "literature search", or "research to support decisions".
iterative-analysis
Performs sequential multi-round oracle analysis where each round builds on previous findings. Use when user specifies analysis count (e.g., "analyze 5 times"), asks to think repeatedly, or needs progressive refinement.
cardiology-trial-editorial
Identify landmark cardiology trials and write evidence-based editorials in Eric Topol's authoritative Ground Truths style. Use when the user wants to: (1) Discover and evaluate recent important trials from top cardiology journals (NEJM, JACC, Lancet, EHJ, Circulation), (2) Assess trial importance using systematic scoring, (3) Write 500-word editorials on cardiology/interventional cardiology advances for physician audiences, (4) Create thought leadership content that demonstrates deep domain expertise. Supports both full-text and abstract-only scenarios with PubMed integration for references.
常见问题
什么是 Claude Skills(Agent Skills)?
Skills 是一种标准化的 SKILL.md 指令包——把「何时触发、如何执行、使用哪些工具」写成结构化工作流,装入 Claude Code、Codex、Cursor 等 AI 编程助手后,Agent 遇到对应任务会自动调用,效果远好于每次手写提示词。
装了 Skill 之后怎么使用?
安装后无需手动调用:直接在 Claude Code / Cursor 里描述你的任务(如「帮我做 PR 代码审查」),Agent 检测到匹配的技能后会自动加载对应 SKILL.md 并按其流程执行。也可以在对话中显式要求「使用 xx skill」。
国内如何高速下载安装?
本站在 Cloudflare R2 托管了全部技能 ZIP 离线包,无需直连 GitHub:在详情页点击「Download ZIP」,或用 `npx skills add <owner/repo>` 一键安装;也可以复制 curl 命令直接解压到 ~/.claude/skills 目录。