OpenClaw 最大的开源医疗人工智能技能库。

这是什么？

OpenClaw Medical Skills 是一个精心策划的 869 项 AI 代理技能合集，涵盖生物医学和临床研究的全方位领域。这些技能专为OpenClaw / NanoClaw——基于Claude的个人AI助手框架——设计，将通用AI代理转变为强大的医学和科学研究伙伴。

每个技能都是一个自包含的模块（文件），其内容包括：SKILL.md

  • 教授代理专业知识和工作流程
  • 连接真实数据库、API和计算工具
  • 产出结构化、临床或科学相关的成果

我们受益于开源社区。完整的资源合集请见此处：Awesome LLM Resources

为何这套收藏重要

无技能 与OpenClaw医疗技能合作
关于医学的通用人工智能回应 真实的PubMed / ClinicalTrials.gov / FDA咨询
无生物信息学能力 RNA-seq，scRNA-seq，GWAS，变异呼叫管线
没有毒品情报 ChEMBL、DrugBank、DDI预测、药物警戒
没有临床记录 SOAP笔记、出院摘要、事先授权决定
没有基因组学支持 VCF注释、ACMG分类、PRS计算
无监管指导 FDA，CE标志，IEC 62304，ISO 14971合规

该合集汇聚了12+个开源技能库的技能，涵盖学术研究工具、临床工作流程、监管框架以及前沿的AI驱动蛋白质设计——为您的AI代理提供可媲美专业研究团队的能力。

安装

要求

  • OpenClaw 已安装并运行，或者用 NanoClaw 作为替代方案
  • Git（用于克隆这个仓库）

对于OpenClaw用户

OpenClaw从两个位置加载技能：

优先级 路径 范围
高 <workspace>/skills/ 按工作空间计算（推荐）
低 ~/.openclaw/skills/ 全局，所有代理共享
方法1 — 克隆与复制（推荐）
# Clone this repository
git clone https://github.com/MedClaw-Org/OpenClaw-Medical-Skills.git

# Install to your workspace skills directory
cp -r OpenClaw-Medical-Skills/skills/* <your-workspace>/skills/

# Or install globally (available to all agents)
cp -r OpenClaw-Medical-Skills/skills/* ~/.openclaw/skills/

技能在下一场游戏中会自动获得。不需要重启。

方法2 — ClawHub CLI

如果你用ClawHub注册表，可以从那里搜索并安装单个技能。对于从这个集合批量安装，方法1更快。

npm install -g clawhub
clawhub install <skill-slug>    # install a single skill
clawhub update --all            # update all installed skills
方法3 — 配置额外目录

要永久指向该仓库的克隆副本，请将其添加到：~/.openclaw/openclaw.json

{
  "skills": {
    "load": {
      "extraDirs": ["/path/to/OpenClaw-Medical-Skills/skills"]
    }
  }
}

这样可以挂载整个集合而不复制文件。

方法4 — 仅安装选定技能

选择与你领域相关的技能：

# Example: clinical + drug discovery stack
SKILLS=(
  "clinical-reports"
  "tooluniverse-drug-research"
  "tooluniverse-pharmacovigilance"
  "clinicaltrials-database"
  "biomedical-search"
  "tooluniverse-drug-drug-interaction"
)

for skill in "${SKILLS[@]}"; do
  cp -r OpenClaw-Medical-Skills/skills/$skill ~/.openclaw/skills/
done

给纳米爪用户

NanoClaw 在启动时将技能加载到代理容器中。container/skills/

# Clone and copy into NanoClaw container skills directory
git clone https://github.com/MedClaw-Org/OpenClaw-Medical-Skills.git
cp -r OpenClaw-Medical-Skills/skills/* /path/to/nanoclaw/container/skills/

# Rebuild the container to apply
cd /path/to/nanoclaw
./container/build.sh

验证

安装后，请咨询您的代理人：

<span style="background-color:#f6f8fa"><span style="color:#1f2328"><span style="color:#1f2328"><span style="background-color:#f6f8fa"><code>What medical and clinical skills do you have available?
</code></span></span></span></span>

你的客服应该列出已安装的技能及其能力。

技能概述

类别 伯爵 亮点
通用与核心 10 浏览器/搜索、文档工具和开发者工作流程工具
医学与临床 119 临床报告、CDS、肿瘤学、影像学和医疗人工智能
科学数据库 43 基因组学/蛋白质/药物数据库与生物医学知识检索
生物信息学（GPTOMICS） 239 变异分析、测序质量控制、去离方、通路、单细胞和表观基因组学
组学与计算生物学 59 单细胞/空间、蛋白质组学、化学信息学和蛋白质设计工具
ClawBio 管道 21 scRNA、GWAS、祖源和结构工作流程的编排流程
BioOS 扩展套件 285 肿瘤学、免疫学、临床人工智能及基础设施扩展代理套件
数据科学与工具 93 统计、可视化、自动化、仿真与科学工具
总计 869

目录

通用与核心

医学与临床

科学数据库

生物信息学（GPTOMICS生物-*套件）

组学与计算生物学

ClawBio 管道

BioOS 扩展套件

数据科学与工具

技能列表

🧰 通用与核心

扩展/折叠该类别

通用工具

点击展开技能列表

技能 描述
代理浏览器 浏览网页以完成任何任务——研究主题、阅读文章、使用网页应用、填写表单、截图、提取数据和测试网页。只要浏览器有用就用。
寻找技能 当用户提出诸如“我该如何做X”、“找到适合X的技能”、“是否有技能可以......”或表达扩展能力的兴趣时，帮助用户发现并安装代理技能。
多搜索引擎 多搜索引擎集成，包含17个搜索引擎（8个CN + 9个全球搜索引擎）。支持百度、必应、360、索国、微信、谷歌、DuckDuckGo、WolframAlpha等。支持高级、时间筛选、网站搜索。不需要API密钥。
维基百科搜索 使用MediaWiki API从维基百科搜索和检索结构化内容，获取可靠且百科全书式的信息。支持多语言查询。
深度研究 对任何主题进行自主多步深度研究。检索多个来源，阅读完整内容，综合发现，并生成结构化报告。用于综合研究、文献综述、竞争分析或主题深入探讨。
PDF 全面的PDF工具包——提取文本和表格，创建新PDF，合并/拆分文档，处理表单，OCR扫描PDF。处理任何.pdf文件时均可使用。
DOCX 创建、编辑和分析Word文档（.docx）。支持跟踪变更、注释、格式保持和文本提取。用于起草、红线划定或从Word文件中提取内容。
XLX 电子表格的创建、编辑和分析。支持公式、格式化、数据分析和可视化。可用于任何.xlsx、.xlsm、.csv或.tsv任务。
PPTX 演示文稿的制作、编辑和分析。支持版面设计、演讲说明、模板和设计。任何.pptx文件都可使用。
文档合著 引导用户通过结构化的文档共同编写工作流程。用于撰写文档、提案、技术规范、决策文档或类似结构化内容。

🏥 医学与临床

扩展/折叠该类别

医疗工具

点击展开技能列表

技能 描述
pubmed-search（pubmed-search） 在PubMed上搜索科学文献。当用户请求查找论文、检索文献、查找研究、查找出版物或询问最新研究时，使用该功能。
医学研究工具包 查询14+生物医学数据库，用于药物再利用、靶点发现、临床试验和文献研究。通过统一的MCP端点访问ChEMBL、PubMed、ClinicalTrials.gov、OpenTargets、OpenFDA、OMIM、Reactome、KEGG、UniProt等。
医学专科简报 为任何医学专科生成每日或按需的医学研究简报。检索顶级期刊（NEJM、柳叶刀、美国医学杂志、英国医学杂志、自然医学杂志）的最新研究，提供简明扼要的摘要，包含一句话的要点和直接链接。当用户询问医学新闻、研究更新或专科更新（内分泌学、心脏病学、肿瘤学、神经学等）时使用。
USMLE 通过进度跟踪、薄弱领域分析、题库管理和住院医师匹配规划，准备美国医学执照考试。涵盖第一步/第二步的CK/第三步、IMG专属指导、分数预测以及身心健康支持。
医疗实体提取器 从患者信息中提取医疗实体（症状、药物、化验值、诊断）。
耐心的AI。 简化患者的医疗文件。收集医生的信件、检查结果、处方、出院摘要和临床记录，并用清晰、个性化的语言进行解释。
生物医学检索 完整的生物医学信息搜索，结合PubMed、预印本、临床试验和FDA药品标签。由Valyu语义搜索驱动。
医学影像审查 撰写医学影像AI研究的综合文献综述。用于撰写关于影像主题的调查论文、系统综述或文献分析。
FHIR-开发-技能 FHIR API开发指南，用于构建医疗端点（患者、观察、就诊、状况、用药请求）。用于开发或集成FHIR REST API。
临床-试验-方案-技能 制定医疗器械或药物的临床试验方案。用于设计临床研究、创建FDA提交文件或制定试验产品方案时。
PRIOR-AUTH-REVIEW-SKILL 自动审核付款方对事前授权（PA）请求。评估医疗必要性，核实保险政策，并做出事先授权决定。
临床报告 撰写全面的临床报告——病例报告（CARE指南）、诊断报告（放射/病理/化验）、临床试验报告（ICH-E3、CSR）和患者文档（SOAP、H&P、出院摘要）。符合HIPAA/FDA/ICH-GCP标准。
临床试验数据库 通过API v2查询 ClinicalTrials.gov。按疾病、药物、地点、状态或阶段搜索试验。通过NCT ID检索试验详情，导出临床研究数据及患者匹配。
临床决策支持 生成用于药物和临床研究的临床决策支持（CDS）文件——患者队列分析、带有GRADE证据分级的治疗建议报告、生物标志物整合以及统计输出（风险比、生存曲线）。
工具宇宙-临床-试验-设计 战略性临床试验设计可行性评估。评估患者群体规模、生物标志物流行率、终点选择、对照分析、安全性监测及监管通路。用于规划1/2期试验或评估试验可行性。
工具宇宙-疾病-研究 生成涵盖流行病学、机制、诊断、治疗及正在进行的试验的综合疾病研究报告。用于询问疾病、综合征或需要系统性疾病分析时。
工具宇宙-文献-深度研究 深入的文献研究，包括目标消歧义、证据分级和结构化主题提取。解析基因/蛋白质鉴定，识别同义词，综合生物学模型，并生成可检验的假设。用于详尽的文献综述或目标画像。
工具宇宙-临床-指南 从12+个来源（NICE、WHO、ADA、AHA/ACC、NCCN、SIGN、CPIC 等）搜索并检索临床实践指南。涵盖心脏病学、肿瘤学、糖尿病、药物基因组学等。在询问治疗建议或标准护理时使用。
工具宇宙-药物-研究 涵盖鉴定、药理学、靶点、临床试验、安全性、药物基因组学和ADMET的综合药物研究报告。用于药物分析、安全性评估或临床开发研究。
工具宇宙-药物再利用 利用基于靶点、化合物和疾病驱动的策略识别药物再利用候选药物。通过分析靶点、生物活性和安全性特征，发现批准药物的新适应症。
工具：宇宙-药物-药物相互作用 药物间相互作用预测与风险评估。分析CYP450/转运蛋白机制、严重程度分类，并提供管理策略。支持多药分析（3+药物）及替代药物推荐。
工具宇宙-罕见病-诊断 基于表型和遗传数据的罕见病鉴别诊断。将症状与HPO术语匹配，识别Orphanet/OMIM中的候选疾病，并解释意义不明的变异。
Tooluniverse-药物警戒 分析FDA不良事件报告、标签警告和药物基因组数据中的药物安全信号。计算PRR/ROR，识别严重不良事件，并评估药物基因组风险。
工具宇宙-临床-试验匹配 精准医疗和肿瘤学的患者与试验匹配。根据分子资格、临床标准、生物标志物比对性和地理可行性对 ClinicalTrials.gov 试验进行排名，并以定量试验匹配评分（0-100）。
文献综述 跨多个数据库（PubMed、arXiv、bioRxiv、Semantic Scholar）的系统文献综述。制作专业格式化的报告，引用经验证，采用APA、Nature、Vancouver格式。
tooluniverse-precision-oncology 基于分子特征的癌症患者可行治疗建议。解读肿瘤突变，识别FDA批准的疗法、临床试验和耐药机制。
工具宇宙-癌症变异-解读 癌症体细胞突变的临床解读。给定基因+变异（例如，EGFR L858R，BRAF V600E），评估致癌性、治疗意义及试验资格。
工具宇宙变体分析 生产准备的VCF处理、变异标注和突变分析。解析VCF文件，使用ClinVar/gnomAD/COSMIC注释，并解释临床意义。
工具宇宙变体解释 从原始呼叫到ACMG分类建议的系统性临床变异解读。汇总来自ClinVar、gnomAD、文献和人口数据库的证据。
tooluniverse-structural-variant-分析 临床基因组学的综合结构变异（SV/CNV）分析。分类SV，评估致病性，并解读拷贝数变化。
tooluniverse-polygenic-risk-score 利用GWAS汇总统计构建并解释复杂疾病的多基因风险评分（PRS）。计算遗传风险特征并解读PRS百分位数。
工具宇宙-精准-医学-分层 通过整合基因组学、临床和治疗数据，实现精准医疗的患者分层。识别与治疗相关的亚组及生物标志物驱动的治疗选项。
工具宇宙-基因-特性-基因 利用GWAS目录（500k+关联）和开放目标遗传学基因座到基因的预测，发现与疾病和性状相关的基因。
工具宇宙-GWAS-药物发现 将GWAS信号转化为药物靶点和再利用机会。执行基因座到基因的定位、可药性评估及现有药物鉴定。
工具宇宙-GWAS-研究-探索器 比较GWAS研究并评估跨队列的重复性。整合NHGRI-EBI的GWAS目录和开放目标遗传学，进行交叉研究荟萃分析。
Tooluniverse-GWAS-微映射 利用统计精细定位识别并优先排序GWAS位点的因果变异。计算因果变异识别的后验概率和可信集合。
tooluniverse-gwas-snp-interpretation Interpret SNPs from GWAS studies by aggregating evidence from GWAS Catalog, Open Targets Genetics, and ClinVar. Retrieves variant-trait associations and functional annotations.
tooluniverse-phylogenetics Phylogenetics and sequence analysis — alignment processing, evolutionary tree construction, and evolutionary metrics for pathogens or species.
tooluniverse-epigenomics Epigenomics data processing — methylation array analysis (CpG filtering, differential methylation), chromatin accessibility, and histone modification analysis.
tooluniverse-rnaseq-deseq2 使用PyDESeq2进行RNA测序差异表达分析。进行归一化、色散估计、Wald测试、LFC收缩和通路富集。
tooluniverse-单胞 使用扫描扫描的单细胞RNA测序分析。执行质量控制、归一化、主耦分析、UMAP、莱顿聚类、轨迹分析和单元类型注释。
工具宇宙-空间-转录组学 空间转录组学数据分析——绘制组织结构中的基因表达图谱。支持10个Visium、MERFISH、seqFISH和Slide-seq平台。
工具宇宙-空间-组学-分析 空间多组学数据集成的计算分析——空间变量基因、结构域注释和组织解析组学。
Tooluniverse-蛋白质组学分析 质谱蛋白质组学分析——蛋白质定量、差异表达、PTMs及蛋白质间相互作用网络构建。
工具宇宙-代谢组学 代谢组学研究——识别代谢物并搜索数据库（HMDB 220k+代谢物、MetaboLights、代谢组学工作台）。
工具宇宙-代谢组学分析 代谢组学数据分析——从LC-MS、GC-MS或NMR数据中进行代谢物鉴定、定量、通路分析及代谢通量。
工具宇宙-多组学-集成 整合转录组学、蛋白质组学、表观基因组学、基因组学和代谢组学，应用于系统生物学和精准医疗。
工具宇宙-多组-疾病-表征 系统层面疾病表征，整合基因组学、转录组学、蛋白质组学、通路和治疗层。
tooluniverse-expression-data-retrieval 检索ArrayExpress和BioStudies中的基因表达和组学数据集，提供高质量评估和结构化报告。
工具宇宙-基因富集 利用gseapy、PANTHER、STRING、Reactome进行基因富集与途径分析。支持GO丰富化、KEGG通路和40+ ToolUniverse工具。
工具宇宙-系统-生物学 系统生物学及途径分析，使用Reactome、KEGG、WikiPathways、Pathway Commons和BioModels。网络建模与通路仿真。
工具宇宙-蛋白质-相互作用 利用STRING、BioGRID和SASBDB进行蛋白质-蛋白质相互作用网络分析。映射交互网络，附有置信度评分和功能模块。
工具宇宙-蛋白质-结构-检索 检索RCSB PDB、PDBe和AlphaFold的蛋白质结构数据，提供质量评估和全面的结构图谱。
工具宇宙-蛋白质-治疗-设计 利用AI引导的新设计设计新型蛋白质治疗药物（结合剂、酶、支架）——包括RFdiffusion、ProteinMPNN和ESM。
工具宇宙-抗体工程 治疗药物的抗体工程与优化——人源化、亲和力成熟、可发展性评估和免疫原性预测。
工具宇宙-免疫-曲目-分析 从测序数据中分析TCR/BCR的组合——克隆性、多样性、V（D）J基因使用、克隆扩增和抗原特异性预测。
工具宇宙-免疫疗法-反应-预测 利用多生物标志物整合——TMB、MSI、PD-L1、TIL特征和HLA分型——预测患者对免疫检查点抑制剂的反应。
工具宇宙-传染病 病原体特性分析及药物在传染病爆发中的再利用。识别分类学、必需蛋白、结构靶点及治疗方案。
tooluniverse-CRISPR-屏幕分析 功能基因组学的CRISPR筛选分析——通过汇集或阵列筛选（敲除/激活/干扰）来识别关键基因和命中。
tooluniverse-target-research Comprehensive biological target intelligence — protein info, structure, interactions, pathways, expression, variant landscape, and drug pipeline.
tooluniverse-network-pharmacology Compound-target-disease network analysis for drug repurposing, polypharmacology discovery, and systems pharmacology.
tooluniverse-statistical-modeling Statistical modeling on biomedical datasets — linear/logistic regression, mixed-effects models, survival analysis, and Bayesian methods.
tooluniverse-image-analysis Biomedical microscopy image analysis — colony morphometry, cell counting, fluorescence quantification, and statistical comparison of imaging data.
literature-search Comprehensive scientific literature search across PubMed, arXiv, bioRxiv, medRxiv using natural language queries powered by Valyu semantic search.
medrxiv-search Search medRxiv medical preprints with natural language queries powered by Valyu semantic search.
clinical-trials-search Search ClinicalTrials.gov with natural language queries — find trials by condition, enrollment status, and outcomes via Valyu.
drug-discovery-search End-to-end drug discovery platform combining ChEMBL, DrugBank, targets, and FDA labels via natural language Valyu search.
drug-labels-search Search FDA drug labels with natural language queries — indications, dosing, and safety data via Valyu.
chembl-search Search ChEMBL bioactive molecules database — compounds, assay data, and bioactivity via Valyu semantic search.
open-targets-search Search Open Targets drug-disease associations and target validation via Valyu semantic search.
patents-search Search global patents with natural language queries — prior art, patent landscapes, and innovation tracking via Valyu.
drugbank-search Search DrugBank comprehensive drug database — mechanisms, interactions, and safety data via Valyu semantic search.
arxiv-search Search arXiv preprints (biology, medicine, AI) using natural language queries powered by Valyu semantic search.
gwas-database Query NHGRI-EBI GWAS Catalog for SNP-trait associations by rs ID, disease/trait, or gene. Retrieve p-values and summary statistics for genetic epidemiology.
scikit-survival Survival analysis and time-to-event modeling in Python — Kaplan-Meier, Cox regression, log-rank tests, and censored data handling using scikit-survival.

药物安全与化学生物学

点击展开技能列表

技能 描述
Tooluniverse-不良事件检测 利用FDA FAERS数据、药品标签、不成比例分析（PRR、ROR、IC）及生物医学证据检测和分析不良药物事件信号。生成定量安全信号评分（0-100）。
工具宇宙-装订器-发现 利用基于结构和配体的方法发现用于蛋白质靶点的新型小分子结合剂。创建包含候选化合物、ADMET配置文件和合成可行性的可作报告。
工具宇宙-化学-化合物-检索 从PubChem和ChEMBL检索化合物信息，并进行消歧义、交叉引用和质量评估。包含标识符、性质、生物活性的综合化合物谱。
工具宇宙-化学-安全 综合化学安全与毒理评估，整合ADMET-AI预测、CTD毒理基因组学、FDA标签安全数据、DrugBank安全档案及STITCH化学-蛋白质相互作用。
工具宇宙-药物-靶点-验证 在10个维度上计算验证药物靶点：消歧义、疾病关联、可药性、化学物质、临床先例、安全性和表达证据。
工具宇宙-序列-检索 从NCBI和ENA检索生物序列（DNA、RNA、蛋白质），并进行基因消歧、登录类型处理和全面的序列谱。

医学影像与病理学

点击展开技能列表

技能 描述
派迪康 用于处理DICOM医学影像文件的Python库。读取、写入、修改DICOM数据、提取像素数据、处理元数据和多帧文件。
历史人物 数字病理学图像处理工具包，用于整张幻灯片图像（WSI）。处理H&E或IHC染色组织图像，提取千像素幻灯片上的瓷砖。
pathml 用于分析 WSI 和多参数成像数据的计算病理学工具包。H&E染色图像、多重免疫荧光、空间组学整合。
omero积分 显微镜数据管理平台。通过Python访问图像，检索数据集，分析像素，管理投资回报率/注释，满足高内容筛选工作流程。
神经套件2 综合生物信号处理：心电图、脑电图、EDA、RSP、PPG、肌电图、心电图信号。心血管信号分析、神经生理学和生理数据处理。
神经像素分析 神经像素神经记录分析。加载SpikeGLX/OpenEphys数据，Kilosort4尖峰排序，质量指标，Allen/IBL策划，用于神经科学研究。

医疗机器学习与临床人工智能

点击展开技能列表

技能 描述
Pyhealth 用于开发基于临床数据（电子健康记录、声明）的机器学习模型的综合医疗人工智能工具包。任务定义API、模型训练、临床自然语言处理评估和预测。
Scikit-Learn Python 中的机器学习：监督学习（分类、回归）、无监督学习（聚类、降维）、模型评估、超参数调优。
变压器 用于自然语言处理、计算机视觉、音频和多模态任务的预训练变压器模型。文本生成、分类、问答和生物医学自然语言处理（BioBERT，ClinicalBERT）。
SHAP 利用SHAP（SHapley加法解释）进行模型可解释性。解释机器学习模型预测，计算特征重要性，生成生物医学模型的SHAP图。
umap-learn UMAP 降维。用于二维/三维可视化、聚类预处理（HDBSCAN）的快速非线性流形学习，适用于高维组学数据。

健康与养生分析

点击展开技能列表

技能 描述
营养分析仪 全面的营养分析：宏量/微量营养素追踪、饮食评估、餐食计划、食品数据查询和营养建议。
心理健康分析仪 心理健康数据分析：情绪追踪、症状模式、PHQ/GAD评分、行为洞察和健康建议。
睡眠分析仪 睡眠质量分析：睡眠阶段、持续时间、效率指标、昼夜节律评估及睡眠卫生建议。
康复分析器 康复进展追踪：功能评估、锻炼计划、康复里程碑及物理/职业治疗的结局测量。
体能分析器 健身表现分析：运动记录、力量/有氧指标、训练负荷、VO2max估计和周期规划。
健康趋势分析仪 纵向健康趋势分析：生命体征追踪、生物标志物趋势、风险因素监测及预测性健康洞察。
减重分析仪 体重管理分析：热量平衡、体成分追踪、进展监测及循证减肥策略。
进球分析器 健康目标追踪与分析：SMART目标设定、进展指标、习惯形成及健康目标的激励见解。
职业健康分析仪 职业健康评估：工作场所人体工学、暴露风险、与工作相关的疾病监测以及重返工作计划。
旅行健康分析器 旅行医学：目的地健康风险、疫苗接种要求、疟疾预防、高山反应及旅行者健康准备。
家庭健康分析仪 家庭健康管理：儿科里程碑、家族病史、预防性筛查计划以及多代健康追踪。
TCM-宪法分析仪 中医体质分析：中医体型评估、型态区分、草药建议及生活方式指导。
紧急卡 制作包含关键健康数据、药物、过敏和紧急联系人的紧急医疗信息卡，以保障患者安全。
ai-analyzer AI-powered comprehensive health data interpretation combining multiple biomarkers and health metrics for holistic wellness assessment.
wellally-tech Technical framework for WellAlly health analytics platform: integration patterns, data pipelines, and health AI infrastructure.

心理健康与危机干预

点击展开技能列表

技能 描述
危机-检测-干预-人工智能 利用自然语言处理和心理健康情绪分析检测危机信号。为心理健康应用和康复平台实施自杀意念检测、自动升级及危机资源集成。
危机应对协议 安全处理心理健康危机情况：危机检测、安全协议、紧急升级、自杀预防以及AI辅导应用的热线整合。
HIPAA合规性 处理PHI时确保符合HIPAA要求。审计日志、数据访问控制、安全事件跟踪以及健康数据应用的合规验证。
临床诊断推理 通过系统性错误分析、鉴别诊断框架和临床判断力提升，识别并抵消医疗决策中的认知偏差。
言语病理学AI。 人工智能驱动的言语语言病理学：音素分析、发音可视化、声音障碍评估、流利度干预、辅助沟通和口吃治疗支持。
HRV-情感失语症专家 心率变异性生物识别和情绪意识训练。HRV分析、内感受训练、生物反馈、迷走神经张力评估和自主神经系统评估。
ADHD-每日计划本 针对多动症优化的日常计划：时间盲友好的排班、执行功能支持、多巴胺感知任务设计以及神经多样性友好的生产力系统。
悲伤伴侣 富有同情心的丧亲支持、纪念的建立、悲伤教育以及通过非线性失落路径的疗愈之旅指导。
jungian-psychologist Jungian analytical psychology: shadow work, archetypal analysis, dream interpretation, active imagination, addiction/recovery through depth psychology lens, and individuation process.
modern-drug-rehab-computer Comprehensive addiction recovery knowledge system: evidence-based treatment (CBT, DBT, MI, EMDR, MAT), recovery resources, crisis intervention, and family systems for rehab environments.
recovery-community-moderator Trauma-informed AI moderation for addiction recovery communities: harm reduction, 12-step traditions, conflict detection, and crisis post identification.

医疗器械与监管

点击展开技能列表

技能 描述
ISO-13485认证 医疗器械 ISO 13485 QMS 文档的综合工具包：间隙分析、质量手册、程序、医疗器械文件。涵盖FDA质量管理、欧盟MDR合规。

🗂️ 科学数据库

扩展/折叠该类别

科学数据库（基因组学与变异）

点击展开技能列表

技能 描述
Clinvar数据库 查询NCBI ClinVar以了解其临床意义的变异。可按基因/位置搜索，解释致病性分类，通过电子工具API或FTP访问，注释VCF，用于基因组医学。
clinpgx-database 访问ClinPGx药物基因组学数据（PharmGKB的继任者）。查询基因-药物相互作用、CPIC指南、等位基因功能，以获取精准医疗和基因型引导剂量决策。
宇宙数据库 访问COSMIC癌症突变数据库。查询体细胞突变、癌症基因普查、突变特征、基因融合，用于癌症研究和精准肿瘤学。需要身份验证。
ensembl-database 查询 Ensembl 基因组数据库 REST API for 250+ 物种。基因查找、序列检索、变异分析、比较基因组学、直系同源分析、VEP预测，以及基因组研究。
gene-database Query NCBI Gene via E-utilities/Datasets API. Search by symbol/ID, retrieve gene info (RefSeqs, GO, locations, phenotypes), batch lookups, for gene annotation and functional analysis.
geo-database Access NCBI GEO for gene expression/genomics data. Search/download microarray and RNA-seq datasets (GSE, GSM, GPL), retrieve SOFT/Matrix files, for transcriptomics and expression analysis.
ena-database Access European Nucleotide Archive via API/FTP. Retrieve DNA/RNA sequences, raw reads (FASTQ), genome assemblies by accession, for genomics and bioinformatics pipelines.
gget CLI/Python toolkit for rapid bioinformatics queries with access to 20+ databases: Ensembl, UniProt, AlphaFold, ARCHS4, Enrichr, OpenTargets, COSMIC, BLAST, and more.
pysam Genomic file toolkit. Read/write SAM/BAM/CRAM alignments, VCF/BCF variants, FASTA/FASTQ sequences, extract regions, calculate coverage, for NGS data processing pipelines.

科学数据库（蛋白质、通路与药物）

点击展开技能列表

技能 描述
alphafold-数据库 访问AlphaFold的2亿兆+人工智能预测蛋白结构。通过UniProt ID检索结构，下载PDB/mmCIF文件，分析置信度量（pLDDT、PAE），用于药物发现和结构生物学。
PDB数据库 访问RCSB PDB以获取3D蛋白质/核酸结构。通过文本/序列/结构搜索，下载坐标（PDB/mmCIF），检索元数据，用于结构生物学和药物发现。
uniprot-database Direct REST API access to UniProt. Protein searches, FASTA retrieval, ID mapping, Swiss-Prot/TrEMBL. For multi-database workflows, prefer bioservices (unified interface to 40+ services).
string-database Query STRING API for protein-protein interactions (59M proteins, 20B interactions). Network analysis, GO/KEGG enrichment, interaction discovery, 5000+ species, for systems biology.
kegg-database Direct REST API access to KEGG (academic use). Pathway analysis, gene-pathway mapping, metabolic pathways, drug interactions, ID conversion.
reactome-database Query Reactome REST API for pathway analysis, enrichment, gene-pathway mapping, disease pathways, molecular interactions, expression analysis, for systems biology.
brenda-database Access BRENDA enzyme database via SOAP API. Retrieve kinetic parameters (Km, kcat), reaction equations, organism data, substrate-specific enzyme info for biochemical research.
hmdb-database Access Human Metabolome Database (220K+ metabolites). Search by name/ID/structure, retrieve chemical properties, biomarker data, NMR/MS spectra, pathways, for metabolomics.
metabolomics-workbench-database Access NIH Metabolomics Workbench via REST API (4,200+ studies). Query metabolites, RefMet nomenclature, MS/NMR data, m/z searches, for metabolomics and biomarker discovery.
pubchem-database Query PubChem via PUG-REST API (110M+ compounds). Search by name/CID/SMILES, retrieve properties, similarity/substructure searches, bioactivity, for cheminformatics.
chembl-database Query ChEMBL's bioactive molecules and drug discovery data. Search compounds by structure/properties, retrieve bioactivity data (IC50, Ki), find inhibitors, for medicinal chemistry.
drugbank-database Access comprehensive drug information from DrugBank including drug properties, interactions, targets, pathways, chemical structures, and pharmacology data.
zinc-database Access ZINC (230M+ purchasable compounds). Search by ZINC ID/SMILES, similarity searches, 3D-ready structures for docking, analog discovery, for virtual screening.
opentargets-database Query Open Targets Platform for target-disease associations, drug target discovery, tractability/safety data, genetics/omics evidence, known drugs, for therapeutic target identification.
fda-database Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis.
pubmed-database Direct REST API access to PubMed. Advanced Boolean/MeSH queries, E-utilities API, batch processing, citation management.
openalex-database Query and analyze scholarly literature using the OpenAlex database. Search for academic papers, analyze research trends, find works by authors or institutions.
biorxiv-database Search bioRxiv preprint server by keywords, authors, date ranges, or categories, retrieving paper metadata for life sciences preprint discovery.
bioservices Primary Python tool for 40+ bioinformatics services. Unified API for UniProt, KEGG, ChEMBL, PubChem, Reactome, QuickGO — preferred for multi-database workflows.
uspto-database Access USPTO APIs for patent/trademark searches, examination history (PEDS), assignments, citations, office actions, for IP analysis and prior art searches.

癌症基因组数据库

点击展开技能列表

技能 描述
cbioportal-database 查询cBioPortal癌症基因组学：体细胞突变、拷贝数、基因表达及数百项癌症研究的生存数据。癌症靶点验证、癌基因分析及患者级基因组分析。
depmap Query the Cancer Dependency Map (DepMap) for cancer cell line gene dependency scores (CRISPR Chronos), drug sensitivity, and gene effect profiles. Identify cancer-specific vulnerabilities and synthetic lethal interactions.
imaging-data-commons Query and download public cancer imaging data from NCI Imaging Data Commons. Access radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required.

基因组与分子数据库

Click to expand skill list

Skill Description
bindingdb-database Query BindingDB for measured drug-target binding affinities (Ki, Kd, IC50, EC50). Drug discovery, lead optimization, polypharmacology, and SAR studies.
gnomad-database Query gnomAD for population allele frequencies, variant constraint scores (pLI, LOEUF), and loss-of-function intolerance. Variant pathogenicity interpretation and rare disease genetics.
gtex-database Query GTEx for tissue-specific gene expression, eQTLs, and sQTLs. Link GWAS variants to gene regulation and interpret non-coding variant effects.
interpro-database Query InterPro for protein family, domain, and functional site annotations. Integrates Pfam, PANTHER, PRINTS, SMART, and 11+ databases for protein function prediction.
jaspar-database Query JASPAR for transcription factor binding site profiles (PWMs/PFMs). Regulatory genomics, motif analysis, and GWAS regulatory variant interpretation.
monarch-database Query the Monarch Initiative knowledge graph for disease-gene-phenotype associations. Integrates OMIM, ORPHANET, HPO, ClinVar for rare disease gene discovery.
tiledbvcf Scalable VCF/BCF ingestion, storage, and parallel queries using TileDB for population genomics at scale.

Structural Biology & Drug Discovery

Click to expand skill list

Skill Description
molecular-dynamics Run and analyze molecular dynamics simulations with OpenMM and MDAnalysis. Protein/small molecule systems, force fields, energy minimization, RMSD/RMSF analysis, free energy surfaces.
glycoengineering Analyze and engineer protein glycosylation. Predict N/O-glycosylation sites, access glycoengineering tools (NetOGlyc, GlycoShield). Therapeutic antibody optimization and vaccine design.
adaptyv Cloud laboratory platform for automated protein testing: binding assays, expression testing, thermostability, enzyme activity. Protein sequence optimization with NetSolP, SoluProt, ESM.
ginkgo-cloud-lab Submit and manage protocols on Ginkgo Bioworks Cloud Lab for autonomous lab execution. Cell-free protein expression, protocol workflows, and biotech automation.

🧬 Bioinformatics (gptomics bio-* suite)

Expand/Collapse this category

Bioinformatics Tools & Pipelines

Click to expand skill list

Skill Description
biopython Primary Python toolkit for molecular biology: PubMed/NCBI queries (Bio.Entrez), sequence manipulation, file parsing (FASTA, GenBank, FASTQ, PDB), BLAST workflows.
scikit-bio Biological data toolkit. Sequence analysis, alignments, phylogenetic trees, diversity metrics (alpha/beta, UniFrac), ordination (PCoA), PERMANOVA, for microbiome analysis.
etetoolkit Phylogenetic tree toolkit (ETE). Tree manipulation (Newick/NHX), evolutionary event detection, orthology/paralogy, NCBI taxonomy, visualization (PDF/SVG), for phylogenomics.
deeptools NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
nextflow-development Run nf-core bioinformatics pipelines (rnaseq, sarek, atacseq) on sequencing data. Use for RNA-seq, WGS/WES, or ATAC-seq from local FASTQs or public datasets (GEO/SRA).
fastq-analysis SRA downloading, FASTQ quality control, STAR alignment, gene quantification, and single-cell kallisto/bustools pipelines for bulk and single-cell sequencing data.
geniml Genomic interval data (BED files) for machine learning tasks. Train region embeddings (Region2Vec, BEDspace), single-cell ATAC-seq analysis.
gtars High-performance genomic interval analysis in Rust with Python bindings. Genomic regions, BED files, coverage tracks, overlap detection, tokenization for ML models.
arboreto Infer gene regulatory networks (GRNs) from gene expression data using GRNBoost2 and GENIE3 algorithms. For bulk RNA-seq and single-cell RNA-seq regulatory network inference.
lamindb Open-source biological data framework for queryable, traceable, reproducible, and FAIR datasets (scRNA-seq, genomics, imaging).
dnanexus-integration DNAnexus cloud genomics platform. Build apps/applets, manage data, dxpy Python SDK, run workflows, FASTQ/BAM/VCF, for genomics pipeline development.
latchbio-integration Latch platform for bioinformatics workflows. Build pipelines with Latch SDK, @workflow/@task decorators, deploy serverless workflows, Nextflow/Snakemake integration.

Bioinformatics — Clinical Databases & Variant Analysis

Click to expand skill list

Skill Description
bio-clinical-databases-clinvar-lookup Query ClinVar for clinical variant classifications, pathogenicity assertions, and review status.
bio-clinical-databases-dbsnp-queries Query dbSNP for SNP frequency, allele, and functional annotation data.
bio-clinical-databases-gnomad-frequencies Retrieve population allele frequencies from gnomAD for rare variant interpretation.
bio-clinical-databases-hla-typing HLA typing from sequencing data using standard typing tools and databases.
bio-clinical-databases-myvariant-queries Batch query MyVariant.info for aggregated variant annotations from multiple databases.
bio-clinical-databases-pharmacogenomics PharmGKB/CPIC pharmacogenomics variant lookup for drug-gene interactions.
bio-clinical-databases-polygenic-risk Calculate polygenic risk scores from GWAS summary statistics and genotype data.
bio-clinical-databases-somatic-signatures Extract and classify mutational signatures from somatic variant catalogs (COSMIC).
bio-clinical-databases-tumor-mutational-burden Compute tumor mutational burden (TMB) from somatic variant calls.
bio-clinical-databases-variant-prioritization Rank and filter candidate variants by pathogenicity scores, inheritance, and phenotype match.
bio-variant-calling GATK-based germline variant calling pipeline from aligned BAM/CRAM files.
bio-variant-calling-clinical-interpretation Interpret variant calls in clinical context with ACMG guidelines.
bio-variant-calling-deepvariant DeepVariant deep-learning variant caller for short-read WGS/WES data.
bio-variant-calling-filtering-best-practices Apply VQSR and hard-filtering best practices to variant call sets.
bio-variant-calling-joint-calling Joint genotyping across multiple samples for improved variant discovery.
bio-variant-calling-structural-variant-calling Call structural variants (SVs) from long-read or paired-end sequencing.
bio-variant-annotation Annotate VCF files with functional, population, and clinical consequence data.
bio-variant-normalization Normalize variant representations (left-alignment, decomposition) for consistent comparison.
bio-vcf-basics Read, write, and parse VCF files; filter by quality, region, and sample.
bio-vcf-manipulation Advanced VCF manipulation: merging, splitting, reformatting, subset extraction.
bio-vcf-statistics Compute variant statistics: ts/tv ratio, heterozygosity, depth distributions.
bio-gatk-variant-calling End-to-end GATK HaplotypeCaller variant calling with BQSR and joint genotyping.
bio-copy-number-cnv-annotation Annotate CNV calls with gene content, database overlap, and clinical significance.
bio-copy-number-cnv-visualization Visualize copy number profiles and segment plots from WGS/WES data.
bio-copy-number-cnvkit-analysis CNVKit copy number analysis for targeted sequencing and WES data.
bio-copy-number-gatk-cnv GATK4 somatic copy number alteration calling pipeline.
bio-tumor-fraction-estimation Estimate tumor purity and ploidy from allele frequencies and copy number data.
bio-ctdna-mutation-detection Detect circulating tumor DNA mutations from liquid biopsy ultra-deep sequencing.
bio-cfdna-preprocessing Process cell-free DNA sequencing data: adapter trimming, deduplication, QC.
bio-methylation-based-detection Detect methylation-based cancer signals from cfDNA methylation data.
bio-longitudinal-monitoring Track somatic variant evolution and clonal dynamics across serial samples.

Bioinformatics — Sequencing & Read QC

Click to expand skill list

Skill Description
bio-fastq-quality Assess FASTQ read quality with FastQC/MultiQC; generate per-sample QC reports.
bio-read-qc-adapter-trimming Trim sequencing adapters with Trimmomatic, Cutadapt, or fastp.
bio-read-qc-contamination-screening Screen reads for human/microbial contamination using FastQ Screen or Kraken.
bio-read-qc-fastp-workflow End-to-end read QC and preprocessing with fastp including UMI handling.
bio-read-qc-quality-filtering Apply quality-score and length filters to remove low-quality reads.
bio-read-qc-quality-reports Aggregate multi-sample QC reports with MultiQC.
bio-read-qc-umi-processing Deduplicate PCR duplicates using UMI-tools for accurate quantification.
bio-paired-end-fastq Handle paired-end FASTQ files: validation, interleaving, splitting.
bio-alignment-io Read/write SAM/BAM/CRAM alignment files with pysam and samtools.
bio-alignment-msa-parsing Parse and analyze multiple sequence alignments (FASTA, ClustalW, Stockholm).
bio-alignment-msa-statistics Compute MSA statistics: conservation, gap content, entropy.
bio-alignment-pairwise Pairwise sequence alignment using Smith-Waterman, Needleman-Wunsch, BLAST.
bio-longread-alignment Align long reads (ONT/PacBio) with minimap2; sort and index BAM files.
bio-longread-qc Quality control for long-read sequencing: read length, N50, error rate.
bio-longread-medaka Consensus polishing and variant calling with Oxford Nanopore Medaka.
bio-longread-structural-variants Call large structural variants from long-read data with Sniffles/PBSV.
bio-basecalling Base-call raw ONT signals with Dorado/Guppy; convert FAST5 to FASTQ.
bio-compressed-files Handle compressed bioinformatics files: bgzip, tabix, zstd, htslib.
bio-format-conversion Convert between bioinformatics formats: FASTQ↔FASTA, BAM↔CRAM, BED↔GTF.
bio-sequence-statistics Compute sequence statistics: GC content, length distributions, complexity.
bio-read-sequences Read and iterate over biological sequences from FASTA/FASTQ files.
bio-write-sequences Write biological sequences to FASTA/FASTQ with metadata preservation.
bio-filter-sequences Filter sequences by length, quality, pattern, or taxonomy label.
bio-batch-processing Batch-process large bioinformatics datasets across samples and cohorts.
bio-rnaseq-qc RNA-seq specific QC: strandedness, rRNA contamination, gene body coverage.
bio-long-read-sequencing-clair3-variants Call variants from long-read sequencing with Clair3 deep-learning model.
bio-long-read-sequencing-isoseq-analysis Iso-Seq full-length transcript analysis for isoform discovery.
bio-long-read-sequencing-nanopore-methylation Call CpG methylation from Oxford Nanopore sequencing with Modbam2bed.
bio-splicing-qc RNA splicing quality assessment: junction read coverage, novel splice sites.
bio-splicing-quantification Quantify alternative splicing events: PSI/inclusion levels per isoform.
bio-sashimi-plots Generate sashimi plots for visualizing RNA-seq splicing at specific loci.
bio-consensus-sequences Generate consensus FASTA sequences by applying VCF variants to a reference using ; useful for sample-specific references and haplotype reconstruction.bcftools consensus

Bioinformatics — Differential Expression & Transcriptomics

Click to expand skill list

Skill Description
bio-de-deseq2-basics DESeq2 differential expression analysis: design matrix, size factors, dispersion.
bio-de-edger-basics EdgeR differential expression for count data with empirical Bayes dispersion.
bio-de-results Extract, filter, and annotate DESeq2/EdgeR results tables.
bio-de-visualization Volcano plots, MA plots, and heatmaps for differential expression results.
bio-differential-expression-batch-correction Remove batch effects with ComBat/limma for multi-cohort DE analysis.
bio-differential-expression-timeseries-de Time-series differential expression with splines and mixed models.
bio-differential-splicing Detect differential alternative splicing events with rMATS or MAJIQ.
bio-isoform-switching Identify isoform switching events with DRIMSeq and IsoformSwitchAnalyzeR.
bio-ribo-seq-orf-detection Detect translated ORFs from ribosome profiling data with RiboTaper/Ribo-TISH.
bio-ribo-seq-riboseq-preprocessing Preprocess ribosome profiling reads: adapter trimming, rRNA removal, alignment.
bio-ribo-seq-ribosome-periodicity Assess triplet periodicity and ribosome footprint quality in Ribo-seq data.
bio-ribo-seq-ribosome-stalling Identify ribosome stalling sites and pausing from Ribo-seq profiles.
bio-ribo-seq-translation-efficiency Compute translation efficiency ratios from matched RNA-seq and Ribo-seq.

Bioinformatics — Pathway & Network Analysis

Click to expand skill list

Skill Description
bio-pathway-go-enrichment Gene Ontology enrichment analysis with clusterProfiler or g:Profiler.
bio-pathway-gsea Gene Set Enrichment Analysis (GSEA) with pre-ranked or count-based statistics.
bio-pathway-kegg-pathways KEGG pathway enrichment and visualization for metabolic/signaling pathways.
bio-pathway-reactome Reactome pathway analysis with hierarchical enrichment and visualization.
bio-pathway-wikipathways WikiPathways enrichment and network visualization.
bio-pathway-enrichment-visualization Dot plots, enrichment maps, and network visualizations for pathway results.

Bioinformatics — Single-Cell & Spatial Omics

Click to expand skill list

Skill Description
bio-single-cell-batch-integration Integrate scRNA-seq datasets across batches with Harmony, BBKNN, scVI.
bio-single-cell-cell-annotation Annotate single-cell clusters using marker genes and reference atlases.
bio-single-cell-cell-communication Infer ligand-receptor cell-cell communication with CellChat or NicheNet.
bio-single-cell-clustering Cluster single cells with Leiden/Louvain algorithms in Scanpy/Seurat.
bio-single-cell-data-io Read/write AnnData, Seurat, and 10x Genomics h5ad/h5 formats.
bio-single-cell-doublet-detection Remove doublets from scRNA-seq with Scrublet or DoubletFinder.
bio-single-cell-lineage-tracing Reconstruct cell lineage trees from scRNA-seq with clonal barcodes.
bio-single-cell-markers-annotation Identify cluster marker genes and auto-annotate cell types.
bio-single-cell-metabolite-communication Infer metabolite-mediated intercellular communication from scRNA-seq.
bio-single-cell-multimodal-integration Integrate scRNA-seq with ATAC, CITE-seq, or spatial using WNN/MultiVI.
bio-single-cell-perturb-seq Analyze genetic perturbation screens from Perturb-seq / CROP-seq data.
bio-single-cell-preprocessing Single-cell preprocessing: count filtering, normalization, HVG selection.
bio-single-cell-scatac-analysis scATAC-seq peak calling, TF motif enrichment, and chromatin accessibility.
bio-single-cell-splicing RNA velocity and splicing dynamics with scVelo or Alevin.
bio-single-cell-trajectory-inference Infer pseudotime trajectories with Monocle3, PAGA, or Slingshot.
bio-spatial-transcriptomics-image-analysis Analyze histology images co-registered with spatial transcriptomics data.
bio-spatial-transcriptomics-spatial-communication Ligand-receptor communication analysis with spatial context (COMMOT, SpatialDE).
bio-spatial-transcriptomics-spatial-data-io Load and process Visium, Slide-seq, MERFISH, and STARmap datasets.
bio-spatial-transcriptomics-spatial-deconvolution Deconvolve cell type proportions in spatial spots with RCTD, SPOTlight.
bio-spatial-transcriptomics-spatial-domains Identify spatially variable genes and tissue domains with SpatialDE/BANKSY.
bio-spatial-transcriptomics-spatial-multiomics Integrate spatial transcriptomics with proteomics, metabolomics, or imaging.
bio-spatial-transcriptomics-spatial-neighbors Build spatial neighbor graphs and perform neighborhood enrichment analysis.
bio-spatial-transcriptomics-spatial-preprocessing Preprocess spatial transcriptomics: QC, normalization, spot filtering.
bio-spatial-transcriptomics-spatial-proteomics Analyze spatial proteomics data from CODEX, IMC, or MIBI platforms.
bio-spatial-transcriptomics-spatial-statistics Spatial statistics: Moran's I, spatial autocorrelation, co-localization.
bio-spatial-transcriptomics-spatial-visualization Visualize spatial gene expression maps and tissue section overlays.

Bioinformatics — Epigenomics & Chromatin

Click to expand skill list

Skill Description
bio-atac-seq-atac-peak-calling Call ATAC-seq chromatin accessibility peaks with MACS2/MACS3.
bio-atac-seq-atac-qc ATAC-seq quality control: TSS enrichment, fragment size, FRiP score.
bio-atac-seq-differential-accessibility Differential chromatin accessibility between conditions with DESeq2/DiffBind.
bio-atac-seq-footprinting Transcription factor footprinting from ATAC-seq with TOBIAS or HINT-ATAC.
bio-atac-seq-motif-deviation TF motif deviation scoring with chromVAR for single-cell ATAC data.
bio-atac-seq-nucleosome-positioning Infer nucleosome positioning from ATAC-seq fragment length distributions.
bio-chipseq-differential-binding Differential ChIP-seq binding analysis with DiffBind.
bio-chipseq-motif-analysis De novo and known motif discovery from ChIP-seq peaks with HOMER/MEME.
bio-chipseq-peak-annotation Annotate ChIP-seq peaks with genomic features and nearest genes.
bio-chipseq-peak-calling Call ChIP-seq peaks with MACS2 for TF binding and histone marks.
bio-chipseq-qc ChIP-seq quality metrics: FRiP, SCC, phantompeakqualtools.
bio-chipseq-super-enhancers Identify super enhancers from H3K27ac ChIP-seq with ROSE.
bio-chipseq-visualization Heatmaps and aggregate profiles at peak regions with deepTools.
bio-hi-c-analysis-compartment-analysis Call A/B compartments from Hi-C contact matrices.
bio-hi-c-analysis-contact-pairs Process Hi-C contact pairs: filtering, deduplication, binning.
bio-hi-c-analysis-hic-data-io Read and write Hi-C data formats: .hic, cool, mcool with cooler/hicstuff.
bio-hi-c-analysis-hic-differential Differential Hi-C interaction analysis between conditions.
bio-hi-c-analysis-hic-visualization Visualize Hi-C contact maps, TADs, and loops with pyGenomeTracks.
bio-hi-c-analysis-loop-calling Detect chromatin loops from Hi-C data with Mustache or HICCUPS.
bio-hi-c-analysis-matrix-operations Normalize Hi-C matrices: ICE, KR, VC; compute observed/expected.
bio-hi-c-analysis-tad-detection Identify topologically associating domains (TADs) from Hi-C data.
bio-methylation-bismark-alignment Align bisulfite sequencing reads and extract CpG methylation with Bismark.
bio-methylation-calling Call CpG methylation from WGBS/RRBS alignments.
bio-methylation-dmr-detection Identify differentially methylated regions (DMRs) with DSS or MethylKit.
bio-methylation-methylkit Methylation analysis with MethylKit: CpG tiles, DMR calling, annotation.

Bioinformatics — Metagenomics & Microbiome

Click to expand skill list

Skill Description
bio-metagenomics-abundance Estimate microbial taxon abundances from shotgun metagenomics.
bio-metagenomics-amr-detection Detect antimicrobial resistance genes with AMRFinder or RGI/CARD.
bio-metagenomics-functional-profiling Functional profiling of metagenomes with HUMAnN3 for pathway/gene families.
bio-metagenomics-kraken Taxonomic classification of metagenomic reads with Kraken2/Bracken.
bio-metagenomics-metaphlan Clade-specific marker-based profiling of microbial communities with MetaPhlAn4.
bio-metagenomics-strain-tracking Track microbial strains across samples with StrainPhlan or inStrain.
bio-metagenomics-visualization Visualize microbiome composition with Krona charts and stacked bar plots.
bio-microbiome-amplicon-processing Process 16S/ITS amplicon sequencing with QIIME2 or DADA2.
bio-microbiome-differential-abundance Test differential microbial abundance with ANCOM-BC, MaAsLin2, or ALDEx2.
bio-microbiome-diversity-analysis Alpha/beta diversity analysis: Shannon, PD, UniFrac, PCoA.
bio-microbiome-functional-prediction Predict functional capacity from 16S data with PICRUSt2 or Tax4Fun.
bio-microbiome-qiime2-workflow End-to-end QIIME2 workflow: denoising, diversity, differential abundance.
bio-microbiome-taxonomy-assignment Assign taxonomy to ASVs/OTUs using SILVA, GTDB, or Greengenes2.

Bioinformatics — Immunoinformatics & Flow Cytometry

Click to expand skill list

Skill Description
bio-immunoinformatics-epitope-prediction Predict MHC-I/II epitopes from protein sequences with NetMHCpan/MHCflurry.
bio-immunoinformatics-immunogenicity-scoring Score peptide immunogenicity for vaccine and neoantigen prioritization.
bio-immunoinformatics-mhc-binding-prediction Predict peptide-MHC binding affinities for multiple alleles.
bio-immunoinformatics-neoantigen-prediction Predict neoantigens from somatic mutations for personalized cancer vaccines.
bio-immunoinformatics-tcr-epitope-binding Predict TCR-epitope binding with ERGO, pMTnet, or NetTCR.
bio-tcr-bcr-analysis-immcantation-analysis Analyze B/T cell receptor repertoires with the Immcantation suite.
bio-tcr-bcr-analysis-mixcr-analysis MiXCR V(D)J alignment and clonotype assembly for immune repertoires.
bio-tcr-bcr-analysis-repertoire-visualization Visualize repertoire diversity, clonal expansion, and V-gene usage.
bio-tcr-bcr-analysis-scirpy-analysis Single-cell TCR/BCR analysis integrated with scRNA-seq using Scirpy.
bio-tcr-bcr-analysis-vdjtools-analysis Immune repertoire statistics and overlap analysis with VDJtools.
bio-flow-cytometry-bead-normalization Normalize flow cytometry data using calibration beads.
bio-flow-cytometry-clustering-phenotyping Cluster and phenotype cell populations with FlowSOM, PhenoGraph, or UMAP.
bio-flow-cytometry-compensation-transformation Apply compensation matrices and biexponential/arcsinh transformations.
bio-flow-cytometry-cytometry-qc Quality control for flow/mass cytometry: signal drift, spillover, outlier detection.
bio-flow-cytometry-differential-analysis Statistical comparison of cell populations between conditions.
bio-flow-cytometry-doublet-detection Detect and remove doublets from flow cytometry data.
bio-flow-cytometry-fcs-handling Read, write, and manipulate FCS files with FlowCore/FlowKit.
bio-flow-cytometry-gating-analysis Manual and algorithmic gating strategies for cell population identification.
bio-imaging-mass-cytometry-cell-segmentation Segment cells in IMC images with Mesmer or CellProfiler.
bio-imaging-mass-cytometry-data-preprocessing Preprocess imaging mass cytometry data: hot pixel removal, normalization.
bio-imaging-mass-cytometry-interactive-annotation Interactively annotate cell types in IMC spatial datasets.
bio-imaging-mass-cytometry-phenotyping Phenotype immune and tumor cells from multi-marker IMC panels.
bio-imaging-mass-cytometry-quality-metrics Quality metrics for IMC acquisitions: signal-to-noise, tissue coverage.
bio-imaging-mass-cytometry-spatial-analysis Spatial cell neighborhood analysis from imaging mass cytometry data.

Bioinformatics — Multi-Omics Integration

Click to expand skill list

Skill Description
bio-multi-omics-data-harmonization Harmonize multi-omics datasets: sample matching, batch correction, feature alignment.
bio-multi-omics-mixomics-analysis Multi-omics factor analysis with mixOmics (DIABLO, MOFA, sPLS-DA).
bio-multi-omics-mofa-integration Multi-Omics Factor Analysis (MOFA+) for latent factor discovery across modalities.
bio-multi-omics-similarity-network Similarity Network Fusion (SNF) for patient stratification from multi-omics.

Bioinformatics — Proteomics & Metabolomics

Click to expand skill list

Skill Description
bio-proteomics-data-import Import DDA/DIA proteomics data from MaxQuant, Proteome Discoverer, FragPipe.
bio-proteomics-dia-analysis DIA proteomics analysis with DIA-NN or Spectronaut.
bio-proteomics-differential-abundance Differential protein abundance with limma, MSstats, or DEqMS.
bio-proteomics-peptide-identification Peptide spectrum matching and database search result parsing.
bio-proteomics-protein-inference Protein grouping, parsimony, and FDR control for proteomics experiments.
bio-proteomics-proteomics-qc Proteomics QC: peptide counts, coverage, missing values, CV.
bio-proteomics-ptm-analysis Post-translational modification analysis: phospho, ubiquitin, glycan enrichment.
bio-proteomics-quantification Label-free, TMT/iTRAQ, and SILAC quantification workflows.
bio-proteomics-spectral-libraries Build and use spectral libraries for DIA data analysis.
bio-metabolomics-lipidomics Lipidomics data analysis: lipid class annotation, fatty acid composition.
bio-metabolomics-metabolite-annotation Annotate mass spec features with HMDB, MZmine, SIRIUS, or MetFrag.
bio-metabolomics-msdial-preprocessing MS-DIAL-based LC-MS/GC-MS data preprocessing and peak detection.
bio-metabolomics-normalization-qc Metabolomics normalization: PQN, LOESS, median, batch correction.
bio-metabolomics-pathway-mapping Map identified metabolites to KEGG, MetaCyc, or Reactome pathways.
bio-metabolomics-statistical-analysis Univariate/multivariate stats for metabolomics: PCA, PLS-DA, ANOVA.
bio-metabolomics-targeted-analysis Targeted metabolomics with MRM/SRM: calibration curves, quantification.
bio-metabolomics-xcms-preprocessing XCMS-based LC-MS peak detection, alignment, and grouping.

Bioinformatics — Structural Biology & Cheminformatics

Click to expand skill list

Skill Description
bio-structural-biology-alphafold-predictions Use AlphaFold2/3 predictions: model quality assessment, confidence scores.
bio-structural-biology-modern-structure-prediction Modern structure prediction with ESMFold, RoseTTAFold, and OpenFold.
bio-pdb-geometric-analysis Geometric analysis of protein structures: distances, angles, contacts, RMSD.
bio-pdb-structure-io Read and write PDB/mmCIF structure files with BioPython or Gemmi.
bio-pdb-structure-modification Modify protein structures: add hydrogens, mutate residues, energy minimize.
bio-pdb-structure-navigation Navigate and inspect PDB structures: chain, residue, atom selection.
bio-molecular-descriptors Calculate molecular descriptors (RDKit): MW, LogP, TPSA, fingerprints.
bio-molecular-io Read/write chemical structure formats: SDF, SMILES, MOL2, PDB with RDKit.
bio-reaction-enumeration Enumerate reactions and products from SMARTS reaction templates.
bio-similarity-searching Molecular similarity search: Tanimoto, fingerprint-based, scaffold hopping.
bio-substructure-search Substructure searching in chemical databases using SMARTS patterns.
bio-virtual-screening Virtual screening workflows: docking, scoring, pose filtering with AutoDock/Vina.
bio-admet-prediction Predict ADMET properties: absorption, distribution, metabolism, excretion, toxicity.

Bioinformatics — Epidemiological & Causal Genomics

Click to expand skill list

Skill Description
bio-epidemiological-genomics-amr-surveillance Antimicrobial resistance surveillance from genomic epidemiology data.
bio-epidemiological-genomics-pathogen-typing Pathogen molecular typing: MLST, wgMLST, cgMLST for outbreak analysis.
bio-epidemiological-genomics-phylodynamics Phylodynamics: molecular clock, population dynamics, BEAST2/TreeTime.
bio-epidemiological-genomics-transmission-inference Infer transmission networks from pathogen genomics with TransPhylo/outbreaker2.
bio-epidemiological-genomics-variant-surveillance Track pathogen variant emergence and spread from genomic surveillance.
bio-causal-genomics-colocalization-analysis Colocalization analysis of GWAS and eQTL signals with coloc or eCAVIAR.
bio-causal-genomics-fine-mapping Fine-map causal variants at GWAS loci with SuSiE or FINEMAP.
bio-causal-genomics-mediation-analysis Causal mediation analysis for gene expression intermediaries.
bio-causal-genomics-mendelian-randomization Two-sample Mendelian randomization with MR-Base/TwoSampleMR.
bio-causal-genomics-pleiotropy-detection Detect horizontal pleiotropy and heterogeneity in MR analyses.
bio-genome-engineering-base-editing-design Design base editors (CBE/ABE) for precise single-base correction.
bio-genome-engineering-grna-design Design and score CRISPR guide RNAs with Cas-OFFinder and CRISPOR.
bio-genome-engineering-hdr-template-design Design HDR templates for precise knock-in via homology-directed repair.
bio-genome-engineering-off-target-prediction Predict CRISPR off-target sites genome-wide for safety assessment.
bio-genome-engineering-prime-editing-design Design pegRNAs and nickase gRNAs for prime editing experiments.
bio-crispr-screens-base-editing-analysis Analyze base editing screens: guide efficiency, editing outcomes.
bio-crispr-screens-batch-correction Correct batch effects in CRISPR screen data across replicates.
bio-crispr-screens-crispresso-editing Quantify editing outcomes with CRISPResso2 from amplicon sequencing.
bio-crispr-screens-hit-calling Call hits from CRISPR screens using MAGeCK, BAGEL2, or casTLE.
bio-crispr-screens-jacks-analysis CRISPR screen analysis with JACKS hierarchical Bayesian model.
bio-crispr-screens-library-design Design CRISPR screen libraries: guide selection, controls, coverage.
bio-crispr-screens-mageck-analysis MAGeCK MLE/RRA analysis for CRISPR pooled screens.
bio-crispr-screens-screen-qc Quality control for CRISPR screens: Gini index, read distribution.

🔬 Omics & Computational Biology

Expand/Collapse this category

Single-Cell & Spatial Omics

Click to expand skill list

Skill Description
anndata Working with annotated data matrices in Python for single-cell genomics analysis, managing experimental measurements with metadata and large-scale omics data.
scanpy Single-cell RNA-seq analysis. Load .h5ad/10X data, QC, normalization, PCA/UMAP/t-SNE, Leiden clustering, marker genes, cell type annotation, trajectory.
scvi-tools Deep learning for single-cell analysis: data integration/batch correction (scVI/scANVI), ATAC-seq (PeakVI), CITE-seq (totalVI), multiome (MultiVI), spatial deconvolution (DestVI).
single-cell-rna-qc Quality control on single-cell RNA-seq data (.h5ad or .h5 files) using scverse best practices with MAD-based filtering and comprehensive visualizations.
cellxgene-census Query CZ CELLxGENE Census (61M+ cells). Filter by cell type/tissue/disease, retrieve expression data, integrate with scanpy/PyTorch, for population-scale single-cell analysis.
pydeseq2 Differential gene expression analysis (Python DESeq2). Identify DE genes from bulk RNA-seq counts, Wald tests, FDR correction, volcano/MA plots.
bulk-combat-correction Remove batch effects from merged bulk RNA-seq or microarray cohorts using pyComBat, with corrected matrix export and pre/post correction visualizations.
bulk-deg-analysis Bulk RNA-seq DEG pipeline: gene ID mapping, DESeq2 normalization, statistical testing, visualization, and pathway enrichment via OmicVerse.
bulk-deseq2-analysis PyDESeq2-based differential expression analysis with ID mapping, DE testing, fold-change thresholding, and enrichment visualization.
bulk-stringdb-ppi Query STRING for protein interactions, build PPI graphs with pyPPI, and render network figures for bulk gene lists.
bulk-to-single-deconvolution Convert bulk RNA-seq cohorts to synthetic single-cell datasets using Bulk2Single workflow for cell fraction estimation and beta-VAE generation.
bulk-trajblend-interpolation Extend scRNA-seq developmental trajectories with BulkTrajBlend by generating intermediate cells from bulk RNA-seq using beta-VAE and GNN models.
bulk-wgcna-analysis Run PyWGCNA through OmicVerse — co-expression module construction, eigengene visualization, and hub gene extraction.
single-annotation Single-cell annotation workflows: SCSA, MetaTiME, CellVote, CellMatch, GPTAnno, and weighted KNN transfer for annotating cell types across modalities.
single-cellphone-db Run CellPhoneDB v5 on annotated single-cell data to infer ligand-receptor networks and produce CellChat-style visualizations.
single-clustering Single-cell clustering workflow: QC, multimethod clustering, topic modeling, cNMF, and cross-batch integration in OmicVerse.
single-downstream-analysis OmicVerse downstream tutorials covering AUCell scoring, metacell DEG, and related exports for single-cell data.
single-multiomics OmicVerse multi-omics tutorials: MOFA, GLUE pairing, SIMBA integration, TOSICA transfer, and StaVIA cartography.
single-preprocessing Single-cell preprocessing in OmicVerse: QC, normalization, HVG detection, PCA/embedding pipelines (CPU/GPU).
single-to-spatial-mapping Map scRNA-seq atlases onto spatial transcriptomics slides using Single2Spatial workflow for deep-forest training and marker visualization.
single-trajectory OmicVerse trajectory workflows: PAGA, Palantir, VIA, velocity coupling, and fate scoring.
spatial-tutorials Spatial transcriptomics tutorials: preprocessing, deconvolution, and downstream modeling across Visium, Visium HD, Stereo-seq, and Slide-seq.
tcga-preprocessing Ingest TCGA sample sheets, expression archives, and clinical carts into OmicVerse, with survival metadata initialization and AnnData export.
gsea-enrichment Gene set enrichment analysis in OmicVerse with correct geneset format handling for loading pathway databases and running GSEA.

Cheminformatics & Drug Discovery

Click to expand skill list

Skill Description
rdkit Cheminformatics toolkit for fine-grained molecular control. SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprints, substructure search, 2D/3D generation, similarity.
datamol Pythonic RDKit wrapper with simplified interface for standard drug discovery: SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformer generation.
medchem Medicinal chemistry filters. Apply drug-likeness rules (Lipinski, Veber), PAINS filters, structural alerts, complexity metrics, for compound prioritization and library filtering.
diffdock Diffusion-based molecular docking. Predict protein-ligand binding poses from PDB/SMILES, confidence scores, virtual screening, for structure-based drug design.
molfeat Molecular featurization for ML (100+ featurizers). ECFP, MACCS, descriptors, pretrained models (ChemBERTa), convert SMILES to features, for QSAR and molecular ML.
deepchem Molecular machine learning toolkit. Property prediction (ADMET, toxicity), GNNs (GCN, MPNN), MoleculeNet benchmarks, pretrained models, for drug discovery ML.
torchdrug Graph-based drug discovery toolkit. Molecular property prediction (ADMET), protein modeling, knowledge graph reasoning, molecular generation, retrosynthesis, GNNs.
torch_geometric Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, molecular property prediction, for geometric deep learning in drug discovery.
pytdc Therapeutics Data Commons. AI-ready drug discovery datasets (ADME, toxicity, DTI), benchmarks, scaffold splits, molecular oracles, for therapeutic ML.
cobrapy Constraint-based metabolic modeling (COBRA). FBA, FVA, gene knockouts, flux sampling, SBML models, for systems biology and metabolic engineering.

Proteomics & Mass Spectrometry

Click to expand skill list

Skill Description
matchms Mass spectrometry spectral analysis. Process mzML/MGF/MSP files, spectral similarity (cosine, modified cosine), metadata harmonization, compound identification.
pyopenms Python interface to OpenMS for LC-MS/MS proteomics and metabolomics workflows. File handling (mzML, mzXML, mzTab, pepXML, mzIdentML) and quantification.
flowio Parse FCS (Flow Cytometry Standard) files v2.0-3.1. Extract events as NumPy arrays, read metadata/channels, convert to CSV/DataFrame, for flow cytometry data preprocessing.

Protein Structure & Design

Click to expand skill list

Skill Description
esm ESM3 generative multimodal protein design (sequence, structure, function) and ESM C efficient protein embeddings. Protein language models for sequence scoring and embedding.
alphafold Validate protein designs using AlphaFold2 structure prediction. Validates designed sequences, predicts binder-target complex structures, calculates pLDDT/PAE metrics.
boltz Structure prediction using Boltz-1/Boltz-2, an open biomolecular structure predictor for protein complexes, binder validation, and open-source AlphaFold alternative.
boltzgen All-atom protein design using BoltzGen diffusion model. Side-chain aware design from the start, designing around small molecules or ligands.
chai Structure prediction using Chai-1 foundation model for protein-protein complexes, binder validation, and protein-small molecule interaction prediction.
rfdiffusion Generate protein backbones using RFdiffusion diffusion model for de novo protein structure generation and binder scaffold design.
bindcraft End-to-end binder design using BindCraft hallucination with built-in AF2 validation for production-quality binder campaigns.
binder-design Guidance for choosing the right protein binder design tool (BoltzGen, BindCraft, or RFdiffusion) and planning binder design campaigns.
proteinmpnn Design protein sequences using ProteinMPNN inverse folding for RFdiffusion backbones, sequence redesign, and partial fixed-position design.
ligandmpnn Ligand-aware protein sequence design using LigandMPNN for sequences around small molecules, enzyme active site design, and binding pocket optimization.
solublempnn Solubility-optimized protein sequence design using SolubleMPNN for E. coli expression, reducing aggregation, and solubility optimization.
foldseek Structure similarity search with Foldseek for finding similar structures in PDB/AFDB databases, structural homology search, and evolutionary relationship discovery.
ipsae Binder design ranking using ipSAE (interprotein Score from Aligned Errors) for ranking binder designs and filtering BindCraft or RFdiffusion outputs.
pdb Fetch and analyze protein structures from RCSB PDB by PDB ID, search for similar structures, prepare targets for binder design.
protein-design-workflow End-to-end guidance for protein design pipelines from project initiation to experimental validation.
protein-qc Quality control metrics and filtering thresholds for protein design: pLDDT, PAE, ipTM for binding, expression, and structure evaluation.
cell-free-expression Guidance for cell-free protein synthesis (CFPS) optimization, troubleshooting low yield/aggregation, and optimizing DNA template design.
binding-characterization Guidance for SPR and BLI binding characterization experiments, kinetics interpretation, and troubleshooting poor binding signal.

Single-Cell & Trajectory Analysis

Click to expand skill list

Skill Description
scvelo RNA velocity analysis. Estimate cell state transitions from unspliced/spliced mRNA dynamics, infer trajectory directions, compute latent time, and identify driver genes in scRNA-seq data.

Phylogenetics & Network Analysis

Click to expand skill list

Skill Description
phylogenetics Build and analyze phylogenetic trees using MAFFT, IQ-TREE 2, and FastTree. Evolutionary analysis, microbial genomics, viral phylodynamics, and molecular clock studies.
networkx Network and graph analysis in Python. Biological network analysis, protein interaction networks, pathway graphs, community detection, and centrality measures.
torch-geometric Graph Neural Networks (PyG) for molecular property prediction, drug-target interaction modeling, and geometric deep learning on biological graphs.

⚙️ ClawBio Pipelines

Expand/Collapse this category

Bioinformatics Orchestration & Pipelines (ClawBio)

Click to expand skill list

Skill Description
bio-orchestrator Meta-agent routing bioinformatics requests to specialized sub-skills. Handles file type detection (VCF, FASTQ, BAM, PDB, h5ad), analysis planning, report generation, and reproducibility export.
scrna-orchestrator Local Scanpy pipeline for single-cell RNA-seq QC, clustering, marker discovery, and two-group differential expression from raw-count .h5ad files.
seq-wrangler Sequence QC, alignment, and BAM processing. Wraps FastQC, BWA/Bowtie2, SAMtools for automated read-to-BAM pipelines.
vcf-annotator Annotate VCF variants with VEP, ClinVar, gnomAD frequencies, and ancestry-aware context. Generates prioritized variant reports.
repro-enforcer Export bioinformatics analyses as reproducible bundles with Conda environment, Singularity container definition, and Nextflow pipeline.
galaxy-bridge Galaxy tool discovery, recommendation, and execution — 8,000+ bioinformatics tools from usegalaxy.org with multi-signal scoring and workflow suggestions.

Genomics, Ancestry & Pharmacogenomics (ClawBio)

Click to expand skill list

Skill Description
gwas-lookup Federated variant lookup across 9 genomic databases: GWAS Catalog, Open Targets, PheWeb (UKB, FinnGen, BBJ), GTEx, eQTL Catalogue, and more.
gwas-prs Calculate polygenic risk scores from DTC genetic data (23andMe/AncestryDNA) using the PGS Catalog.
pharmgx-reporter Pharmacogenomic report from DTC genetic data — 12 genes, 31 SNPs, 51 drugs with CPIC guidelines and personalized dosage cards.
clinpgx Query the ClinPGx API for pharmacogenomic gene-drug data, clinical annotations, CPIC guidelines, and FDA drug labels.
drug-photo Identify a medication from a packaging photo via Claude vision, then retrieve genotype-informed dosage guidance.
claw-ancestry-pca Ancestry decomposition PCA against the Simons Genome Diversity Project (345 samples, 164 global populations).
genome-compare Compare genome to reference individuals and estimate ancestry composition via IBS and EM admixture.
equity-scorer Compute HEIM diversity and equity metrics from VCF or ancestry data. Generates heterozygosity, FST, PCA plots, and HEIM Equity Score with markdown reports.
claw-metagenomics Shotgun metagenomics profiling: taxonomy (Kraken2/Bracken), resistome (CARD/RGI), and functional pathways (HUMAnN3) from paired-end FASTQ.
ukb-navigator Semantic search across UK Biobank's 12,000+ data fields and publications — find the right variables for your research question.

Structural Biology & Literature (ClawBio)

Click to expand skill list

Skill Description
struct-predictor Local protein structure prediction with AlphaFold, Boltz, or Chai. Compare structures, compute RMSD, visualize 3D models.
lit-synthesizer Search PubMed and bioRxiv, summarize papers with LLM, build citation graphs, and generate literature review sections.
claw-semantic-sim Semantic Similarity Index for disease research literature using PubMedBERT embeddings. Compute research equity metrics (HEIM).
labstep Interact with the Labstep electronic lab notebook API. Query experiments, protocols, resources, and inventory.
profile-report Generate structured bioinformatics analysis profile reports.

🧠 BioOS Extended Suite

Expand/Collapse this category

BioOS Extended Bioinformatics Suite (mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills-)

Sequence & Alignment Tools

Click to expand skill list

Skill Description
bio-alignment-sorting Sort SAM/BAM files by coordinate or name with samtools sort.
bio-alignment-filtering Filter alignments by flag, quality, region, or paired status.
bio-alignment-indexing Index BAM/CRAM files with samtools index for random access.
bio-alignment-validation Validate alignment file integrity and detect truncated/corrupt files.
bio-alignment-files-bam-statistics Compute alignment statistics: flagstat, idxstats, coverage depth.
bio-sam-bam-basics Read, inspect, and manipulate SAM/BAM files with samtools/pysam.
bio-duplicate-handling Mark and remove PCR duplicates with Picard or samtools markdup.
bio-pileup-generation Generate base-level pileup from BAM for variant calling and coverage.
bio-reference-operations Download, index, and manage reference genome FASTA files.
bio-blast-searches Run BLAST searches against local or remote databases for sequence homology.
bio-local-blast Set up and run BLAST+ locally with custom databases.
bio-entrez-search Search NCBI Entrez databases (PubMed, gene, nucleotide, protein, SRA).
bio-entrez-fetch Fetch records from NCBI Entrez by accession or UID.
bio-entrez-link Retrieve linked records across NCBI Entrez databases.
bio-uniprot-access Query UniProt for protein sequences, annotations, and cross-references.
bio-geo-data Download and parse GEO datasets and series matrices.
bio-sra-data Download raw sequencing data from NCBI SRA with fasterq-dump.
bio-batch-downloads Batch download bioinformatics data from NCBI, EBI, Ensembl.
Sequence Analysis

Click to expand skill list

Skill Description
bio-seq-objects Work with BioPython sequence objects: SeqRecord, features, annotations.
bio-sequence-properties Compute sequence properties: MW, pI, hydrophobicity, extinction coefficient.
bio-sequence-similarity Compute sequence similarity with pairwise alignment and percent identity.
bio-sequence-slicing Slice, extract, and manipulate subsequences from FASTA/FASTQ.
bio-motif-search Search sequences for regulatory motifs using FIMO, MAST, or regex.
bio-codon-usage Analyze codon usage bias and optimize sequences for expression.
bio-transcription-translation Transcribe and translate DNA sequences; handle genetic code variations.
bio-reverse-complement Compute reverse complement and strand-aware sequence operations.
bio-primer-design-primer-basics Design PCR primers with Primer3 for standard amplification.
bio-primer-design-primer-validation Validate primer specificity by BLAST and thermodynamic analysis.
bio-primer-design-qpcr-primers Design qPCR/RT-PCR primers with efficiency and specificity optimization.
bio-restriction-sites Find restriction enzyme recognition sites in DNA sequences.
bio-restriction-mapping Create restriction maps and in silico digestion patterns.
bio-restriction-fragment-analysis Analyze restriction fragment patterns for cloning and gel prediction.
bio-restriction-enzyme-selection Select restriction enzymes for cloning based on cut sites and compatibility.
Read Alignment

Click to expand skill list

Skill Description
bio-read-alignment-bwa-alignment Align short reads to reference genome with BWA-MEM.
bio-read-alignment-bowtie2-alignment Align short reads with Bowtie2; local and end-to-end modes.
bio-read-alignment-hisat2-alignment Splice-aware RNA-seq alignment with HISAT2.
bio-read-alignment-star-alignment High-speed STAR aligner for RNA-seq with junction detection.
Genome Assembly

Click to expand skill list

Skill Description
bio-genome-assembly-long-read-assembly De novo assembly from ONT/PacBio long reads with Flye or Canu.
bio-genome-assembly-hifi-assembly HiFi (CCS) read assembly with Hifiasm for high-accuracy genomes.
bio-genome-assembly-short-read-assembly Illumina de novo assembly with SPAdes for metagenomes/bacteria/transcriptomes.
bio-genome-assembly-metagenome-assembly Metagenomic assembly: co-assembly, binning, MAG recovery.
bio-genome-assembly-assembly-qc Assess assembly quality with QUAST, BUSCO, and NGA50 metrics.
bio-genome-assembly-assembly-polishing Polish assemblies with Medaka (ONT) or NextPolish (Illumina).
bio-genome-assembly-scaffolding Scaffold contigs with Hi-C, optical mapping, or long reads.
bio-genome-assembly-contamination-detection Detect and remove contamination in assembled genomes.
Genome Intervals & Annotation

Click to expand skill list

Skill Description
bio-genome-intervals-bed-file-basics Read, write, and filter BED files with pybedtools/bedtools.
bio-genome-intervals-interval-arithmetic Intersect, subtract, merge, and complement genomic intervals.
bio-genome-intervals-proximity-operations Find nearest features and compute distances between intervals.
bio-genome-intervals-coverage-analysis Compute read depth coverage across genomic regions.
bio-genome-intervals-bigwig-tracks Create and query BigWig signal tracks from BAM/bedGraph.
bio-genome-intervals-gtf-gff-handling Parse and manipulate GTF/GFF annotation files.
bio-bedgraph-handling Process bedGraph coverage files: arithmetic, normalization, conversion.
RNA Quantification

Click to expand skill list

Skill Description
bio-rna-quantification-featurecounts-counting Count reads per gene with featureCounts from subread package.
bio-rna-quantification-alignment-free-quant Pseudo-alignment quantification with Salmon or Kallisto.
bio-rna-quantification-tximport-workflow Import Salmon/Kallisto quantification into R/DESeq2 with tximport.
bio-rna-quantification-count-matrix-qc QC count matrices: library size, zero inflation, gene detection rates.
bio-expression-matrix-counts-ingest Load and validate count matrices from multiple quantification tools.
bio-expression-matrix-gene-id-mapping Map between Ensembl, Entrez, HGNC, and gene symbol identifiers.
bio-expression-matrix-metadata-joins Join sample metadata to expression matrices for downstream analysis.
bio-expression-matrix-sparse-handling Handle sparse count matrices efficiently with scipy sparse formats.
Epitranscriptomics & CLIP-seq

Click to expand skill list

Skill Description
bio-epitranscriptomics-merip-preprocessing Preprocess MeRIP-seq data for m6A methylation analysis.
bio-epitranscriptomics-m6a-peak-calling Call m6A peaks from MeRIP-seq with exomePeak2 or MACS2.
bio-epitranscriptomics-m6anet-analysis Nanopore direct RNA m6A detection with m6Anet deep learning.
bio-epitranscriptomics-m6a-differential Differential m6A methylation analysis between conditions.
bio-epitranscriptomics-modification-visualization Visualize RNA modification profiles and metagene plots.
bio-clip-seq-clip-preprocessing Preprocess CLIP-seq/eCLIP data: adapter trimming, demultiplexing.
bio-clip-seq-clip-alignment Align CLIP-seq reads with STAR; handle unique mappers.
bio-clip-seq-clip-peak-calling Call RBP binding peaks from CLIP-seq with PureCLIP or MACS2.
bio-clip-seq-binding-site-annotation Annotate CLIP-seq peaks with genomic features and RNA regions.
bio-clip-seq-clip-motif-analysis Discover RBP binding motifs from CLIP-seq peak sequences.
Small RNA-seq

Click to expand skill list

Skill Description
bio-small-rna-seq-smrna-preprocessing Preprocess small RNA-seq: adapter trimming, size selection.
bio-small-rna-seq-mirdeep2-analysis Identify and quantify known/novel miRNAs with miRDeep2.
bio-small-rna-seq-mirge3-analysis miRNA annotation and quantification with miRge3.0.
bio-small-rna-seq-target-prediction Predict miRNA target genes with TargetScan or miRDB.
bio-small-rna-seq-differential-mirna Differential miRNA expression analysis with DESeq2/edgeR.
Population Genetics & Phasing

Click to expand skill list

Skill Description
bio-population-genetics-plink-basics PLINK2 for GWAS QC, LD pruning, and basic population genetics.
bio-population-genetics-population-structure Population stratification with PCA, ADMIXTURE, and STRUCTURE.
bio-population-genetics-linkage-disequilibrium Compute LD metrics (r², D') and LD decay analysis.
bio-population-genetics-association-testing GWAS association testing with PLINK, BOLT-LMM, or SAIGE.
bio-population-genetics-scikit-allel-analysis Population genetics analysis with scikit-allel: diversity, Fst, haplotypes.
bio-population-genetics-selection-statistics Detect natural selection signatures: iHS, XP-EHH, Tajima's D.
bio-phasing-imputation-haplotype-phasing Phase variants with SHAPEIT4 or BEAGLE.
bio-phasing-imputation-genotype-imputation Impute missing genotypes using Michigan/TOPMed imputation servers.
bio-phasing-imputation-reference-panels Select and prepare reference panels (1KGP, HRC, TOPMed) for imputation.
bio-phasing-imputation-imputation-qc QC imputed data: R² filter, INFO score, allele concordance.
Comparative Genomics & Phylogenetics

Click to expand skill list

Skill Description
bio-comparative-genomics-ortholog-inference Infer orthologs and paralogs with OrthoFinder or OMA.
bio-comparative-genomics-synteny-analysis Detect syntenic blocks between genomes with MCScan or SyRI.
bio-comparative-genomics-positive-selection Test for positive selection with PAML, HyPhy, or dN/dS ratios.
bio-comparative-genomics-hgt-detection Detect horizontal gene transfer events in microbial genomes.
bio-comparative-genomics-ancestral-reconstruction Reconstruct ancestral sequences and traits with ASR methods.
bio-phylo-tree-io Read/write phylogenetic trees in Newick, Nexus, PhyloXML formats.
bio-phylo-modern-tree-inference Maximum likelihood tree inference with IQ-TREE 2 or FastTree.
bio-phylo-tree-manipulation Root, prune, reorder, and annotate phylogenetic trees.
bio-phylo-tree-visualization Visualize trees with iTOL, ETE3, or ggtree.
bio-phylo-distance-calculations Compute pairwise phylogenetic distances and diversity metrics.
Systems Biology & Metabolic Modeling

Click to expand skill list

Skill Description
bio-systems-biology-flux-balance-analysis Flux balance analysis (FBA) with COBRApy for metabolic network modeling.
bio-systems-biology-metabolic-reconstruction Reconstruct genome-scale metabolic models from genome annotations.
bio-systems-biology-gene-essentiality Predict essential genes by single gene knockouts in metabolic models.
bio-systems-biology-context-specific-models Build context-specific metabolic models from expression data (GIMME, iMAT).
bio-systems-biology-model-curation Curate SBML metabolic models: mass/charge balance, gap filling.
Experimental Design & Reporting

Click to expand skill list

Skill Description
bio-experimental-design-sample-size Power analysis and sample size calculation for omics experiments.
bio-experimental-design-power-analysis Statistical power analysis for detecting differential signals.
bio-experimental-design-batch-design Optimize sample batching to minimize confounding with ComBat design.
bio-experimental-design-multiple-testing Multiple testing correction: Bonferroni, BH/FDR, q-values.
bio-machine-learning-omics-classifiers Train classifiers on omics data: random forest, SVM, XGBoost.
bio-machine-learning-biomarker-discovery Identify biomarkers from omics data with LASSO, elastic net, SHAP.
bio-machine-learning-model-validation Cross-validation, AUC-ROC, calibration, and permutation testing.
bio-machine-learning-survival-analysis Survival ML: RSF, DeepSurv, CoxBoost from omics features.
bio-machine-learning-atlas-mapping Map query cells to reference atlases with scANVI or Symphony.
bio-machine-learning-prediction-explanation Explain omics ML predictions with SHAP and feature importance.
bio-reporting-automated-qc-reports Generate automated MultiQC-style reports for omics pipelines.
bio-reporting-jupyter-reports Create Jupyter notebook reports with reproducible analysis code.
bio-reporting-rmarkdown-reports Render Rmarkdown reports with integrated plots and statistics.
bio-reporting-quarto-reports Build Quarto multi-format reports (HTML/PDF) from analysis code.
bio-reporting-figure-export Export publication-quality figures in PDF/SVG/TIFF at specified DPI.
bio-research-tools-biomarker-signature-studio Build, validate, and visualize multi-omic biomarker signatures.
End-to-End Workflow Pipelines

Click to expand skill list

Skill Description
bio-workflows-fastq-to-variants Complete FASTQ → alignment → variant calling pipeline.
bio-workflows-rnaseq-to-de RNA-seq → alignment → counts → DESeq2 differential expression.
bio-workflows-scrnaseq-pipeline Single-cell RNA-seq end-to-end: Cell Ranger → Scanpy → clustering.
bio-workflows-atacseq-pipeline ATAC-seq: trimming → alignment → peak calling → differential.
bio-workflows-chipseq-pipeline ChIP-seq: alignment → peak calling → motif analysis → annotation.
bio-workflows-methylation-pipeline WGBS/RRBS: bismark alignment → methylation calling → DMR detection.
bio-workflows-metagenomics-pipeline Metagenomics: QC → classification → functional profiling → AMR.
bio-workflows-metabolomics-pipeline LC-MS/GC-MS: preprocessing → annotation → statistical analysis.
bio-workflows-proteomics-pipeline DDA/DIA proteomics: search → quantification → differential abundance.
bio-workflows-gwas-pipeline GWAS: QC → imputation → association → fine-mapping → annotation.
bio-workflows-somatic-variant-pipeline Tumor-normal somatic variant calling with GATK Mutect2/Strelka2.
bio-workflows-cnv-pipeline Copy number variant detection: WGS/WES CNV calling and annotation.
bio-workflows-spatial-pipeline Spatial transcriptomics: alignment → deconvolution → domain detection.
bio-workflows-multi-omics-pipeline Multi-omics integration pipeline: MOFA, SNF, similarity network fusion.
bio-workflows-multiome-pipeline 10x Multiome: joint scRNA-seq + scATAC-seq processing and integration.
bio-workflows-hic-pipeline Hi-C contact map generation, normalization, TAD/loop calling.
bio-workflows-neoantigen-pipeline Neoantigen prediction: somatic variants → MHC binding → immunogenicity.
生物-工作流程-微生物组-管道 微生物组：16S/ITS扩增型或散弹式，→多样性→差异。
bio-workflows-crispr-screen-pipeline CRISPR界面：引导计数→MAGeCK→点击，调用→可视化。
bio-workflows-CRISPR-编辑-管道 CRISPR编辑：扩增子测序→CRISPResso2→结局分析。
bio-workflows-tcr-pipeline TCR/BCR：V（D）J比对→克隆型→词汇分析。
bio-workflows-riboseq-pipeline Ribo-seq：ORF检测→周期性→足迹对齐。
bio-workflows-smrna-pipeline 小RNA-seq：miRNA鉴定→定量→DE分析。
bio-workflows-merip-pipeline MeRIP-seq：m6A峰呼叫→差异→基序分析。
生物-工作流程-剪辑-管道 CLIP-seq：峰值调用→结合位点注释→基序发现。
bio-workflows-imc-pipeline 成像质细胞术：分段→表型→空间分析。
生物-工作流程-细胞术-管道 流/质量细胞术：质量控制（QC）→门控→聚类→差异化。
bio-workflows-longread-sv-pipeline 长读结构变体调用与注释流水线。
生物工作流程-基因组组装-流程 新基因组组装：原始读段→组装→质量控制→注释。
生物-工作流程-爆发-管道 病原体基因组学：组装→分型→系统动力学→传递。
生物-工作流程-生物标志-管道 生物标志物发现：组学→特征选择→验证→报告。
生物-工作流程-代谢建模-管道 代谢模型重建→FBA→模拟→可视化。
生物剪接管道 替代拼接分析：rMATS → PSI →差分→刺身。
生物-液体-活检-管道 液体活检：cfDNA/ctDNA质检 →突变检测→TMB→MRD。
生物-工作流程-管理-snakemake-workflows 创建和管理Snakemake可重复的生物信息学工作流程。
生物-工作流程-管理-Nextflow-pipelines 构建并运行Nextflow（DSL2）生物信息学流程。
生物-工作流程-管理-CWL-工作流 编写通用工作流语言（CWL）可移植的工作流定义。
生物-工作流-管理-WDL-工作流 为Terra/Cromwell生物信息学执行创建WDL工作流程。
生物工作流程表达到通路 从微分表达到Go/KEGG/反应组富集及途径可视化的端到端工作流程。
Data Visualization (Bioinformatics)

Click to expand skill list

Skill Description
bio-data-visualization-heatmaps-clustering Hierarchical clustering heatmaps with ComplexHeatmap or seaborn.
bio-data-visualization-volcano-customization Customized volcano plots with ggplot2 or matplotlib for DE results.
bio-data-visualization-circos-plots Circular genome visualization with Circos or pycirclize.
bio-data-visualization-genome-browser-tracks Generate genome browser tracks and IGV sessions from BAM/BigWig.
bio-data-visualization-genome-tracks Multi-panel genome track plots with pyGenomeTracks.
bio-data-visualization-ggplot2-fundamentals R ggplot2 for publication-quality genomics and omics figures.
bio-data-visualization-interactive-visualization Interactive omics visualizations with Plotly, Bokeh, or shiny.
bio-data-visualization-upset-plots UpSet plots for multi-set intersection visualization.
bio-data-visualization-multipanel-figures Compose multipanel publication figures with cowplot or patchwork.
bio-data-visualization-color-palettes Scientific color palettes: colorblind-safe, perceptually uniform, diverging.
bio-data-visualization-specialized-omics-plots Specialized plots: lollipop (mutations), circomap, oncoprint.

Oncology & Precision Medicine Agents (BioOS)

Click to expand skill list

Skill Description
autonomous-oncology-agent Autonomous oncology research agent: literature mining, trial matching, biomarker analysis, and treatment hypothesis generation.
precision-oncology-agent Precision oncology: tumor molecular profiling → actionable alterations → treatment recommendations.
pan-cancer-multiomics-agent Pan-cancer multi-omics integration for cross-cancer pattern discovery and driver identification.
tumor-clonal-evolution-agent Model tumor clonal evolution: phylogenetic trees, clonal dynamics, branching patterns from somatic variants.
tumor-heterogeneity-agent Analyze intratumoral heterogeneity from bulk and single-cell sequencing data.
tumor-mutational-burden-agent Compute TMB and assess its predictive value for immunotherapy response.
chromosomal-instability-agent Quantify chromosomal instability (CIN) from copy number and SV data.
cancer-metabolism-agent Analyze tumor metabolic reprogramming from transcriptomic and metabolomic data.
liquid-biopsy-analytics-agent Comprehensive liquid biopsy analytics: ctDNA detection, MRD monitoring, treatment response.
ctdna-dynamics-mrd-agent Track ctDNA dynamics for minimal residual disease detection and treatment monitoring.
MRD-边缘-检测-代理 通过深度测序实现超灵敏MRD检测，并带有错误抑制。
HRD分析代理 同源重组缺乏（HRD）分析用于预测PARP抑制剂反应。
计算病理代理 计算病理学：系统性结构分析、组织分割、组织学特征提取。
多模态-辐射路径-聚变-代理 融合放射学和病理影像学进行综合癌症表型分析。
放射学-病态-聚变-剂 提取放射性和病理特征并进行整合以进行预测建模。
RADGPT-放射报告员 AI辅助放射学报告从影像结果生成。
类器官药物反应剂 分析患者来源类器官中的药物反应以预测个性化治疗。
PDX-模型-分析-代理 患者来源的异种移植模型分析，用于药物疗效和生物标志物发现。
深度视觉蛋白质组学代理 深视觉蛋白质组学：基于激光捕捉显微解剖MS数据的空间蛋白质组分析。
外泌体EV分析剂 细胞外囊泡与外泌体分析：货物剖析与生物标志物发现。
微生物组-癌症-病原 肿瘤微生物组分析及其在癌症进展和免疫治疗反应中的作用。
生物片段分析 分析cfDNA片段大小分布及片段组学特征（FinaleToolkit/Griffin），用于癌症检测和组织起源评估。

Hematology & Blood Disorders (BioOS)

Click to expand skill list

Skill Description
myeloma-mrd-agent Multiple myeloma MRD assessment from flow cytometry and NGS data.
mpn-progression-monitor-agent Myeloproliferative neoplasm progression monitoring from serial molecular data.
mpn-research-assistant Research assistant for myeloproliferative neoplasms: literature, mutation analysis, treatment.
bone-marrow-ai-agent Bone marrow analysis: blast counting, immunophenotyping, disease classification.
hemoglobinopathy-analysis-agent Hemoglobin variant analysis, sickle cell, and thalassemia genotype-phenotype assessment.
chip-clonal-hematopoiesis-agent Clonal hematopoiesis of indeterminate potential (CHIP) variant detection and risk assessment.
coagulation-thrombosis-agent Coagulation pathway analysis, thrombophilia assessment, anticoagulation guidance.

Immunology & Cell Therapy (BioOS)

Click to expand skill list

Skill Description
cart-design-optimizer-agent Optimize CAR-T cell construct design: scFv selection, linker, co-stimulatory domain.
armored-cart-design-agent Design armored CAR-T cells with cytokine payloads and resistance mechanisms.
tcell-exhaustion-analysis-agent Analyze T cell exhaustion from scRNA-seq and ATAC-seq data.
nk-cell-therapy-agent NK cell therapy design: receptor engineering, expansion protocols, persistence.
tcr-pmhc-prediction-agent Predict TCR-pMHC binding affinity and selectivity for TCR therapy design.
tcr-repertoire-analysis-agent TCR repertoire analysis: V(D)J usage, clonotype dynamics, antigen specificity.
immune-checkpoint-combination-agent Predict optimal immune checkpoint combination strategies from tumor immune microenvironment.
tme-immune-profiling-agent Tumor microenvironment immune profiling: cell type deconvolution and spatial mapping.
cytokine-storm-analysis-agent 细胞因子风暴检测、严重程度评分和干预建模。

Single-Cell & Spatial Agents (BioOS)

Click to expand skill list

Skill Description
cellagent-annotation AI-driven single-cell cluster annotation using marker gene databases.
universal-single-cell-annotator Universal scRNA-seq annotator using foundation models and multi-reference integration.
scfoundation-model-agent Single-cell foundation model inference (scFoundation/scGPT) for zero-shot annotation.
rna-velocity-agent RNA velocity analysis with scVelo for trajectory and fate decision inference.
spatial-transcriptomics-agent End-to-end spatial transcriptomics analysis: QC, deconvolution, domain detection.
spatial-transcriptomics-analysis Spatial transcriptomics analysis with Squidpy and SpatialDE.
spatial-agent Spatial omics agent: integrate spatial data with imaging, protein, and genomic layers.
生态位形成者-空间-代理 利用组织微环境的空间生态位分析，采用生态位形成基础模型。
空间表观基因组学代理 空间表观基因组学分析：空间解析的染色质可及性与基因调控。
生物信息学-单胞 通用单细胞生物信息学：聚类、轨迹、细胞通讯。
SCRNA-QC 单细胞RNA测序质量控制：双重态去除、环境RNA、过滤阈值。
Compbioagent-Explorer 计算生物学探索代理，用于多组学数据集分析。
SIMO-多组-集成-代理 单细胞多组学集成与SIMO/MOFA+进行关节嵌入。
表观基因组学-甲基-GPT因子 采用MethylGPT启发的方法进行表观基因组学和DNA甲基化分析。
生物主工作流程 BioMaster 用于端到端生物信息学分析的工作流程编排。

Drug Discovery & Design (BioOS)

Click to expand skill list

Skill Description
agentd-drug-discovery AgentD autonomous drug discovery: target identification, hit finding, ADMET optimization.
chematagent-drug-discovery CheMatAgent: chemistry-aware drug design with retrosynthesis and property optimization.
chemcrow-drug-discovery ChemCrow drug discovery toolkit: web search, Python, chemical tools integration.
medea-therapeutic-discovery MEDEA therapeutic discovery: multimodal evidence aggregation for target-disease validation.
molecule-evolution-agent 定向分子进化：化合物优化与文库设计的生成模型。
分子胶水发现剂 分子胶水发现：诱导邻近降解剂和三元复合物稳定剂。
Protac-设计代理 PROTAC设计：E3连接酶配体选择、连接单元优化、三元复形建模。
TPD-三元-复杂-代理 靶向蛋白质降解三元复合物建模与协同预测。
法师抗体发生器 MAGE antibody generator: sequence design, humanization, affinity maturation.
antibody-design-agent Antibody design: epitope mapping, CDR engineering, bispecific construction.
aav-vector-design-agent AAV vector design: capsid selection, promoter optimization, payload capacity.
protein-structure-prediction Protein structure prediction with AlphaFold3, ESMFold, or Boltz with comparison.
crispr-guide-design CRISPR guide RNA design with on-target scoring and off-target minimization.
crispr-offtarget-predictor Predict CRISPR Cas9/Cas12 off-target sites genome-wide with CRISPOR/Cas-OFFinder.
chemical-property-lookup Look up chemical properties from PubChem, ChEMBL, DrugBank by name/SMILES.
chemistry-agent 用于合成规划、反应预测和性质计算的通用化学试剂。
冷冻-人工智能-药物设计-剂 基于低温电磁结构的人工智能引导药物设计：结合位点分析与对接。
时间解析冷冻剂 动态结构生物学中的时间分辨冷冻电磁分析。
CNV 呼叫代理 专门的CNV检测代理，集成多来电者与集合评分。
Popeve变异预测因子 利用基于EVE群体的进化模型进行变异致病性预测。
变异性-致病性 结构和进化编码变异的VARADD致病性评分。
变体解释-ACMG 基于证据的分类框架下ACMG/AMP变异解读。
基因-面板-设计-代理 设计用于临床或研究测序应用的靶向基因面板。
药物基因组学-代理 药物基因组学分析：变异药物相互作用预测与剂量建议。
多重祖源-PRS-代理 多祖源多基因风险评分计算，并加祖源特异权重。
PRS-Net-深度学习-代理 利用PRSnet进行复杂性状的深度学习PRS预测。
细胞自由RNA代理 游细胞RNA分析：用于液体活检诊断的血浆cfRNA剖析。
长读序列试剂 长读长测序分析：SV呼叫、甲基化、同构体发现、组装。
贝叶斯优化器 《贝叶斯优化用于实验设计和生物医学研究中超参数调优》。

Clinical AI & Healthcare (BioOS)

Click to expand skill list

Skill Description
chatehr-clinician-assistant EHR clinical assistant: note summarization, structured data extraction, clinical decision support.
clinical-note-summarization Summarize clinical notes into structured SOAP format with key findings.
clinical-nlp-extractor Extract clinical entities (diagnoses, medications, procedures) from unstructured text.
ehr-fhir-integration EHR-FHIR integration: HL7 FHIR resource creation, querying, and workflow automation.
fhir-development FHIR API development: build SMART on FHIR apps and FHIR resource endpoints.
digital-twin-clinical-agent Create patient digital twins for treatment simulation and outcome prediction.
trial-eligibility-agent Assess patient eligibility for clinical trials from EHR data and trial criteria.
trialgpt-matching TrialGPT patient-to-trial matching with eligibility assessment from clinical notes.
wearable-analysis-agent Analyze wearable sensor data: activity, sleep, HRV, ECG for health monitoring.
multimodal-medical-imaging Multimodal medical imaging analysis: CT, MRI, PET fusion and segmentation.
prior-auth-coworker Prior authorization workflow assistant for insurance approval processes.
care-coordination Care coordination agent: multi-disciplinary team communication and care plan management.
claims-appeals Insurance claims appeals: documentation preparation and denial reasoning analysis.
lab-results Lab result interpretation: reference ranges, trend analysis, critical value alerts.
drug-interaction-checker Check drug-drug interactions from patient medication lists with severity scoring.
regulatory-drafter Draft regulatory submissions: FDA, EMA, ICH document preparation.
监管起草 医疗器械/药品提交的监管写作与文档结构。
生物医学数据分析 全面的生物医学数据分析：统计、可视化与解读。
数据可视化-生物医学 生物医学专用数据可视化：临床试验图、生存曲线、森林图。

Research Infrastructure & Agents (BioOS)

Click to expand skill list

Skill Description
biomni-general-agent BioMni general biomedical agent for flexible multi-step research tasks.
biomni-research-agent BioMni research-focused agent with literature, database, and analysis integration.
biokernel BioKernel: unified computational kernel for bioinformatics tool orchestration.
biomcp-server BioMCP: Model Context Protocol server for bioinformatics tool access.
mcpmed-bioinformatics-server MCP server providing medical bioinformatics tool access to agents.
kragen-knowledge-graph KRAGEN knowledge graph for biomedical entity relationships and reasoning.
leads-literature-mining LEADS literature mining: automated extraction of biological findings from papers.
knowledge-synthesis Synthesize knowledge from multiple biomedical sources into structured summaries.
deep-research-swarm Multi-agent swarm for deep scientific research with parallel literature synthesis.
research-literature Research literature management: search, organize, and synthesize scientific papers.
search-strategy Design systematic search strategies for scientific literature and databases.
scientific-manuscript Scientific manuscript writing and revision with journal-specific formatting.
cellular-senescence-agent Cellular senescence analysis: marker scoring, SASP profiling, tissue aging assessment.
ngs-analysis Next-generation sequencing data analysis orchestration and QC.
opentrons-protocol-agent Opentrons liquid handler protocol design for automated lab workflows.
虚拟实验室代理 用于计算机模拟和协议优化的虚拟实验室代理。
数据可视化专家 为复杂的科学和临床数据集提供专家数据可视化。
龙虾-生物信息学 通过Lobster AI运行生物信息学分析：scRNA-seq、bulk RNA-seq、文献挖掘、数据集发现、QC和可视化。

📊 Data Science & Tools

Expand/Collapse this category

Statistics & Data Analysis

Click to expand skill list

Skill Description
statistical-analysis Statistical analysis toolkit. Hypothesis tests (t-test, ANOVA, chi-square), regression, correlation, Bayesian stats, power analysis, assumption checks, APA reporting.
statsmodels Statistical modeling: OLS, GLM, logistic, ARIMA, time series, hypothesis tests, diagnostics, AIC/BIC, for rigorous statistical inference.
pymc Bayesian modeling with PyMC. Build hierarchical models, MCMC (NUTS), variational inference, LOO/WAIC comparison, posterior checks, for probabilistic programming.
simpy Process-based discrete-event simulation for clinical systems: queues, resources, time-based events. Useful for modeling hospital workflows and patient flow.
exploratory-data-analysis Comprehensive exploratory data analysis on scientific data files across 200+ file formats — structure, content, quality assessment, and visualization.
data-stats-analysis Statistical tests, hypothesis testing, correlation analysis, and multiple testing corrections using scipy and statsmodels (OmicVerse).
data-transform Transform, clean, reshape, and preprocess biological data using pandas and numpy (OmicVerse).
data-viz-plots 使用 matplotlib 和 seaborn（OmicVerse）创建发表质量的图表和可视化。
科学可视化 用matplotlib/seaborn/plotly创建出版图。多面板布局、错误条、重要性标记、色盲安全、PDF/EPS/TIFF导出。

Lab Automation & Integration

Click to expand skill list

Skill Description
opentrons-integration Lab automation platform for Flex/OT-2 robots. Write Protocol API v2 protocols, liquid handling, hardware modules (heater-shaker, thermocycler), labware management.
pylabrobot Laboratory automation toolkit for controlling liquid handlers, plate readers, pumps, heater shakers, incubators, centrifuges, and analytical equipment.
benchling-integration Benchling R&D platform integration. Access registry (DNA, proteins), inventory, ELN entries, workflows via API, build Benchling Apps, for lab data management automation.
labarchive-integration Electronic lab notebook API integration. Access notebooks, manage entries/attachments, backup notebooks, integrate with Protocols.io/Jupyter/REDCap.
protocolsio-integration Integration with protocols.io API for managing scientific protocols — search, create, update, publish protocols, and manage protocol steps and reagents.
instrument-data-to-allotrope Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format for LIMS systems, data lakes, and downstream analysis.

Scientific Research & Writing

Click to expand skill list

Skill Description
scientific-writing Write scientific manuscripts in full paragraphs using a two-stage process: section outlines then full text. Covers all sections of research papers.
scientific-critical-thinking Evaluate research rigor. Assess methodology, experimental design, statistical validity, biases, confounding, evidence quality (GRADE, Cochrane ROB).
scientific-brainstorming Research ideation partner. Generate hypotheses, explore interdisciplinary connections, challenge assumptions, develop methodologies, identify research gaps.
hypothesis-generation Generate testable hypotheses. Formulate from observations, design experiments, explore competing explanations, develop predictions, propose mechanisms.
科学问题选择 帮助科学家选择研究问题、项目构思、排查卡住的项目以及战略性科学决策。
同行评审 系统同行评审工具包。评估方法论、统计学、设计、可重复性、伦理、数据完整性、报告标准，以供稿件和资助评审。
引用管理 全面的引用管理。在Google Scholar和PubMed搜索论文，提取准确元数据，验证引用，生成BibTeX条目。
研究资助 为国家科学基金会（NSF）、美国国立卫生研究院（NIH）、能源部（DOE）和美国高级研究计划局（DARPA）撰写竞争性研究提案。机构特定的格式、审查标准、预算准备以及更广泛的影响。
研究查找 你可以用 Perplexity 的 Sonar Pro Search 或 OpenRouter 上的 Sonar Reasoning Pro 查找最新研究。自动选择查询复杂度的最佳模型。
生物医学 自主生物医学人工智能代理框架，用于执行涵盖基因组学、药物发现、分子生物学和临床分析的复杂研究任务。
治疗计划 为所有临床专科（包括普通内科、康复、心理健康和慢性病）生成简明（3-4页）的LaTeX/PDF格式医疗治疗计划。

Analyst Personas

Click to expand skill list

Skill Description
biologist-analyst Expert biologist analyst persona for interpreting biological experiments, sequencing data, cell biology assays, and molecular biology research.
chemist-analyst Expert chemist analyst persona for interpreting chemical data, synthesis routes, spectroscopic results, reaction mechanisms, and laboratory analyses.
epidemiologist-analyst Expert epidemiologist analyst persona for study design, cohort analysis, risk factor assessment, public health surveillance, and causal inference.
心理学家-分析师 专家心理学家分析师角色，专注于行为数据分析、心理评估解读、临床案例构建和心理健康研究。

Public Health & Time Series

Click to expand skill list

Skill Description
datacommons-client Access public health statistics from Google Data Commons: disease prevalence, demographic data, health indicators across global sources.
TimesFM预测 与谷歌TimesFM合作的零样本时间序列预测。用于生命体征趋势、健康传感器数据和无定制模型训练的纵向健康监测。
永恒 时间序列机器学习：分类、回归、聚类、异常检测、时间健康数据的分段及连续临床测量。

Scientific Literature & Reference Management

Click to expand skill list

Skill Description
bgpt-论文搜索 使用 BGPT MCP 服务器搜索科学论文。每篇论文返回25+个结构化字段：方法、结果、样本量、质量评分。用于文献综述和证据综合。
皮佐特罗 通过 Zotero Web API v3 以编程方式与 Zotero 参考库交互。检索、创建、更新项目、导出引用、上传PDF，并构建研究自动化工作流程。
打开笔记本 自托管的NotebookLM替代方案。导入PDF、视频、网页、文档;生成AI驱动的笔记;与研究资料交流;支持16+ AI提供商。

Data Processing & Scientific Computing

点击展开技能列表

技能 描述
达斯喀彻来 针对大于RAM的基因组学/组学数据集的分布式计算。扩展pandas/NumPy超越内存，支持并行文件处理，分布式机器学习。
极地 快速内存数据帧库（1-100GB）。更快的pandas替代生物医学数据ETL和分析流程。
Vaex 数十亿行的非核心数据帧作。为大型基因组和临床数据集提供快速统计和可视化。
扎尔-派森 云存储用的是分块的N-D阵列。压缩阵列、并行I/O、S3/GCS集成用于大规模组学数据。
火炬闪电 组织了PyTorch为生物医学人工智能设计的深度学习：多GPU训练、回调、日志记录、临床/基因组模型的分布式训练。

科学可视化与传播

点击展开技能列表

技能 描述
Matplotlib 低层绘图库，可实现完全自定义。科学论文和期刊的出版质量数据。
海生 带熊猫积分的统计可视化。箱型图、小提琴图、热力图、生物医学数据探索的配对图。
情节性 互动可视化。鼠标悬停信息、缩放、仪表盘，用于探索性生物医学分析和演示。
信息图 通过迭代AI优化，创建专业的科学信息图表。支持10种信息图表类型和8种行业风格。
科学示意图 发表级科学图：神经网络架构、生物通路、系统图、流程图。
科学幻灯片 为会议、研讨会、论文答辩制作研究演示幻灯片。支持PowerPoint和LaTeX Beamer。
乳胶海报 用LaTeX制作专业研究海报（beamerposter、tikzposter）。多栏布局的会议海报。
PPTX海报 可导出为PDF或PPTX的HTML/CSS研究海报。现代基于网络的海报设计。
折扣美人鱼写作 科学文档，使用Markdown和24种美人鱼图示类型。9个科学报告文档模板。
Paper-2-web 将学术论文转换为互动网站、演讲视频和会议海报（Paper2Web、Paper2Video、Paper2Poster）。

附加科学工具

点击展开技能列表

技能 描述
派穆 多目标优化，使用PYMOO。药物设计参数优化、帕累托前沿分析、进化算法。
减价 将文档（PDF、DOCX、PPTX、HTML、图片）转换为Markdown以便处理和分析。
困惑搜索 通过Perplexity实现实时科学信息检索的AI驱动搜索。
地百看板 使用 GeoPandas 进行地理空间数据分析。流行病学地图绘制、疾病分布、空间健康分析。
低基因性 自动化假设生成与表格数据集检验。将文献洞察与数据驱动的假设验证相结合。
PDF处理 高级PDF处理：文本提取、表格解析、注释、表单填充。
PDF-处理-Pro 专业的PDF处理，配备增强的OCR、多列布局处理和批量处理。
PDF-人类学 用于科学和医学文档理解的人类化优化PDF分析。
XLX-官方 官方Excel/XLSX技能，用于电子表格的创建、分析和数据管理。
docx-official（文档） 官方Word/DOCX文档创建、编辑和格式化技能。
PPTX官方 官方PowerPoint/PPTX技能，用于演示文稿制作和编辑。

计算仿真与本体论（HeshamFS/材料模拟技能）

点击展开技能列表

技能 描述
本体验证器 验证生物医学本体结构和术语关系（HPO、GO、MeSH、SNOMED、OBO）。
本体探索器 导航和查询生物医学本体：术语层级、注释、交叉引用。
本体映射器 生物医学本体之间的映射：HPO↔OMIM、GO↔UniProt、疾病↔表型交叉本体。
slurm-job-script-generator 生成针对HPC基因组学/生物信息学流水线作业的SLURM批次脚本，并实现优化资源请求。
数值积分 选择并配置生物模型仿真（刚性系统，IMEX）的常微分方程/偏微分方程时间积分。
非线性求解器 配置非线性求解器用于生物网络优化、参数拟合和FBA。
参数优化 实验设计、灵敏度分析、生物模型校准的贝叶斯优化。
线性求解器 选择用于大规模生物网络和代谢模型计算的线性求解器。
数值稳定性 分析时间依赖生物模拟（CFL标准、刚度）的数值稳定性。
仿真编排器 协调多重模拟活动：参数清理、批处理作业、结果聚合。
仿真验证器 验证仿真：预检、运行时监控、收敛、NaN/Inf检测。
收敛研究 用于空间/时间收敛分析，结合理查森外推法进行模拟验证。
后期处理 提取、分析并可视化仿真输出数据：时间序列、现场剖面、统计数据。
性能剖析 识别计算瓶颈，分析扩展行为，优化高性能计算仿真作业。
微分方案 选择有限差分/体积/谱方案以进行生物模型中的偏微分方程离散化。
时间步进 生物动力学的自适应时间步控制：CFL约束，检查点调度。
网格生成 数值仿真网格生成：分辨率、质量指标、自适应细化。

开发者工作流程技能（obra/superpowers）

点击展开技能列表

技能 描述
测试驱动开发 TDD工作流程：在实现前写测试，红绿重构循环以确保代码可靠。
系统调试 结构化调试方法：假设形成、证据收集、根本原因分析。
分派并行代理 为独立任务编排并行子代理以最大化吞吐量。
写作计划 在动用复杂多步骤任务的代码前，先制定结构化的实施计划。
执行计划 在单独的会议中执行书面实施计划，并设置检查点。
头脑风暴 在实施前对需求和设计进行结构化的创新探索。
写作技巧 通过正确的格式和部署验证，创建并验证新的 SKILL.md 技能。
完成前验证 在宣称工作完成前，执行验证命令并确认输出。
请求代码审查 结构化代码审查请求，包含上下文、变更摘要和具体问题。
接收代码审查 流程代码以技术严谨性而非盲目接受的反馈审查。
子代理驱动开发 将开发任务拆分成子任务，以便并行执行子代理。
使用git-worktrees 创建独立的 git 工作树，用于功能工作和计划执行。
完成开发分支 完成开发分支：合并、PR或清理，并有结构化决策选项。
使用超能力 元技能：在对话开始时发现并使用任何任务可用的技能。

致谢

我们受益于以下优秀项目。如果你感兴趣，请去看看。

  1. 超棒的克劳德技能
  2. 人类学技能
  3. Skillsmp
  4. 超棒的克劳德技能
  5. 爪中心
  6. 水产市场
  7. Skills.Sh
  8. 超棒的特工技能
  9. LLMBASE
  10. OpenClaw
  11. 超棒的开爪技能
  12. 克劳德-科学技能
  13. LLMs-通用生命科学与临床技能-
# 人工智能
Logo
AtomGit开源社区

AtomGit 是由开放原子开源基金会联合 CSDN 等生态伙伴共同推出的新一代开源与人工智能协作平台。平台坚持“开放、中立、公益”的理念，把代码托管、模型共享、数据集托管、智能体开发体验和算力服务整合在一起，为开发者提供从开发、训练到部署的一站式体验。

更多推荐

TCP 三次握手与四次挥手

本文深入解析了TCP协议中的三次握手与四次挥手机制，从报文结构、状态机转换到实战抓包分析。三次握手通过SYN、SYN-ACK、ACK报文建立双向连接，确保双方收发能力正常；四次挥手通过FIN、ACK报文独立关闭双向通道。文章还探讨了关键问题：为什么需要三次握手而非两次/四次、随机初始序列号的作用、SYN Flood攻击与防御、CLOSE_WAIT泄漏以及TIME_WAIT状态持续2MSL的原因。通

avatar AtomGit开源社区

Agent Skills 完全指南：AI 编程助手的标准化“能力包”

本文介绍了AI编程助手（如Claude Code）中的Agent Skills机制，这是一种标准化、可复用的"能力包"系统。Skills不同于一次性指令，而是将复杂工作流程（如代码审查、自动部署）封装为可共享的操作手册。文章对比了Skills与Commands、Agents、Hooks的区别，详细解析了Skill的文件结构（SKILL.md为核心）和工作流程，并提供了Spring Boot自动部署

avatar AtomGit开源社区
cover

OpenAI 推出的 GPT-5.5 大模型@ACP#GSV5600应用迭代

avatar AtomGit开源社区