Pick what you need. 只选你用得上的。

How to use the Adaptyv Bio Foundry API and Python SDK for protein experiment design, submission, and results retrieval. Use this skill whenever the user mentions Adaptyv, Foundry API, protein binding assays, protein screening experiments, BLI/SPR assays, thermostability assays, or wants to submit protein sequences for experimental characterization. Also trigger when code imports `adaptyv`, `adaptyv_sdk`, or `FoundryClient`, or references `foundry-api-public.adaptyvbio.com`. 如何使用 Adaptyv Bio Foundry API 和 Python SDK 进行蛋白质实验设计、提交和结果检索。每当用户提及 Adaptyv、Foundry API、蛋白质结合测定、蛋白质筛选实验、BLI/SPR 测定、热稳定性测定或想要提交蛋白质序列进行实验表征时，请使用此技能。当代码导入“adaptyv”、“adaptyv_sdk”或“FoundryClient”或引用“foundry-api-public.adaptyvbio.com”时也会触发。

Predict protein structure for monomers and multimers with AlphaFold2 via the ColabFold runner (Mirdita et al. 2022, github.com/sokrypton/ColabFold; AlphaFold2 Jumper et al. 2021). Reach for this skill to fold a sequence or complex with the AF2/AF2-Multimer evoformer, to validate designed sequences by self-consistency pLDDT, ipTM, and RMSD, or to run a quick MSA-backed prediction using the public MMseqs2 server. 通过 ColabFold 运行器用 AlphaFold2/AF2-Multimer 预测单体与复合物的蛋白质结构（Mirdita 等 2022；Jumper 等 2021）。用于折叠序列或复合物、用 pLDDT/ipTM/RMSD 自洽性验证设计序列，或基于公共 MSA 服务器做快速预测。

Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census. 单细胞分析中带注释矩阵的数据结构。在处理 .h5ad 文件或与 scverse 生态系统集成时使用。这是数据格式技巧——分析工作流程使用scanpy；对于概率模型，使用 scvi-tools；对于人口规模查询，请使用 cellxgene-census。

Infer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Supports distributed computation for large-scale datasets. 使用可扩展算法（GRNBoost2、GENIE3）从基因表达数据推断基因调控网络 (GRN)。在分析转录组学数据（批量 RNA-seq、单细胞 RNA-seq）时使用，以识别转录因子-靶基因关系和调控相互作用。支持大规模数据集的分布式计算。

Search scientific papers and retrieve structured experimental data extracted from full-text studies via the BGPT MCP server. Returns 25+ fields per paper including methods, results, sample sizes, quality scores, and conclusions. Use for literature reviews, evidence synthesis, and finding experimental details not available in abstracts alone. 通过 BGPT MCP 服务器搜索科学论文并检索从全文研究中提取的结构化实验数据。每篇论文返回 25 个以上字段，包括方法、结果、样本量、质量分数和结论。用于文献综述、证据综合以及查找仅在摘要中无法获得的实验细节。

Use this skill when working with Brain Imaging Data Structure (BIDS) datasets: organizing neuroscience and biomedical data (MRI, EEG, MEG, iEEG, PET, microscopy, NIRS, motion capture, EMG, MR spectroscopy, behavioral), querying BIDS layouts, validating compliance, converting DICOM to BIDS, writing metadata sidecars, or creating BIDS derivatives. 处理 Brain Imaging Data Structure (BIDS) 数据集时使用：组织神经科学和生物医学数据（MRI、EEG、MEG、iEEG、PET、显微镜、NIRS、运动捕捉、EMG、磁共振波谱、行为数据），查询 BIDS layout，验证合规性，将 DICOM 转为 BIDS，编写元数据 sidecar，或创建 BIDS derivatives。

Comprehensive molecular biology toolkit. Use for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Best for batch processing, custom bioinformatics pipelines, BLAST automation. For quick lookups use gget; for multi-service integration use bioservices. 综合分子生物学工具包。用于序列操作、文件解析 (FASTA/GenBank/PDB)、系统发育学和程序化 NCBI/PubMed 访问 (Bio.Entrez)。最适合批量处理、定制生物信息学管道、BLAST 自动化。要快速查找，请使用 gget；对于多服务集成，请使用生物服务。

Unified Python interface to 40+ bioinformatics services. Use when querying multiple databases (UniProt, KEGG, ChEMBL, Reactome) in a single workflow with consistent API. Best for cross-database analysis, ID mapping across services. For quick single-database lookups use gget; for sequence/file manipulation use biopython. 与 40 多种生物信息学服务的统一 Python 接口。在具有一致 API 的单个工作流程中查询多个数据库（UniProt、KEGG、ChEMBL、Reactome）时使用。最适合跨数据库分析、跨服务的 ID 映射。对于快速单数据库查找，请使用 gget；对于序列/文件操作，请使用biopython。

Structure prediction for protein, nucleic-acid, and small-molecule complexes with Boltz-2 (Passaro & Wohlwend et al. 2025, github.com/jwohlwend/boltz). Reach for this skill to validate designed binders against a target, to co-fold a protein with a SMILES or CCD ligand, or to get an open-source AlphaFold3 alternative with optional binding-affinity prediction. 用 Boltz-2 预测蛋白质、核酸与小分子复合物结构（Passaro & Wohlwend 等 2025），可选结合亲和力预测。用于验证设计的结合体、与 SMILES/CCD 配体共折叠，或作为开源的 AlphaFold3 替代方案。

Predict genome-wide functional tracks (RNA-seq, CAGE, DNase, ChIP) from DNA sequence with Borzoi. Use this skill when: (1) Scoring the regulatory effect of a variant on expression/accessibility, (2) Generating predicted coverage tracks for a locus, (3) Prioritising non-coding variants by predicted track delta. 用 Borzoi 从 DNA 序列预测全基因组功能轨迹（RNA-seq、CAGE、DNase、ChIP）。用于评估变异对表达/可及性的调控效应、生成某位点的预测覆盖轨迹，或按预测轨迹差异为非编码变异排序。

End-to-end bulk RNA-seq orchestrator — takes raw FASTQ reads through QC and trimming (FastQC, fastp/Trim Galore), alignment and quantification (STAR, Salmon, featureCounts), assembles a gene-level counts matrix, then hands off to differential expression (pydeseq2), pathway/GSEA enrichment (pathway-enrichment), and publication figures (scientific-visualization). Use whenever the user has bulk RNA-seq reads or quant output and wants a complete, reproducible differential-expression workflow — e.g. "analyze my RNA-seq", "FASTQ to DESeq2", "run nf-core/rnaseq", "STAR/Salmon quantification", "build a counts matrix for DESeq2", or "go from reads to differentially expressed genes and enriched pathways". Routes between an nf-core/rnaseq (Nextflow) path and a standalone STAR/Salmon path, and covers experimental design, strandedness, and QC gates. For single-cell RNA-seq use the scanpy skill instead. 端到端 bulk RNA-seq 编排技能：从原始 FASTQ 到 QC/修剪、比对与定量（STAR、Salmon、featureCounts）、基因级 counts 矩阵，再交给差异表达（pydeseq2）、通路/GSEA 富集和发表级图表。用于完整、可复现的 bulk RNA-seq 差异表达工作流；单细胞 RNA-seq 请使用 scanpy。

Query the CZ CELLxGENE Census programmatically for versioned public single-cell and spatial transcriptomics data. Use when you need population-scale cell metadata, gene expression slices, Census summary counts, source H5AD URIs/downloads, embeddings, spatial Census data, or reference atlas comparisons across organisms, tissues, diseases, assays, and cell types. For analyzing your own local single-cell data use scanpy, anndata, or scvi-tools. 以编程方式查询 CELLxGENE 普查（61M+ 单元）。当您需要来自最大的精选单细胞图谱的跨组织、疾病或细胞类型的表达数据时使用。最适合人口规模查询、参考图集比较。要分析您自己的数据，请使用 scanpy 或 scvi-tools。

Structure prediction for protein, nucleic-acid, and small-molecule complexes with the Chai-1 foundation model (Chai Discovery 2024, github.com/chaidiscovery/chai-lab). Reach for this skill to predict an antibody-antigen or protein-ligand complex from a single FASTA, to re-fold designed binders as an AlphaFold-multimer alternative, or to drive co-folding from Python for batched campaigns on a GPU. 用 Chai-1 基础模型预测蛋白质、核酸与小分子复合物结构（Chai Discovery 2024）。用于从单个 FASTA 预测抗体-抗原或蛋白-配体复合物、作为 AlphaFold-multimer 替代重折叠设计的结合体，或用 Python 驱动批量共折叠。

Generate professional clinical decision support (CDS) documents for pharmaceutical and clinical research settings, including patient cohort analyses (biomarker-stratified with outcomes) and treatment recommendation reports (evidence-based guidelines with decision algorithms). Supports GRADE evidence grading, statistical analysis (hazard ratios, survival curves, waterfall plots), biomarker integration, and regulatory compliance. Outputs publication-ready LaTeX/PDF format optimized for drug development, clinical research, and evidence synthesis. 为制药和临床研究环境生成专业的临床决策支持 (CDS) 文件，包括患者队列分析（根据结果进行生物标志物分层）和治疗推荐报告（带有决策算法的循证指南）。支持 GRADE 证据分级、统计分析（风险比、生存曲线、瀑布图）、生物标志物集成和法规遵从性。输出针对药物开发、临床研究和证据合成进行优化的可发布 LaTeX/PDF 格式。

Write comprehensive clinical reports including case reports (CARE guidelines), diagnostic reports (radiology/pathology/lab), clinical trial reports (ICH-E3, SAE, CSR), and patient documentation (SOAP, H&P, discharge summaries). Full support with templates, regulatory compliance (HIPAA, FDA, ICH-GCP), and validation tools. 撰写全面的临床报告，包括病例报告（CARE 指南）、诊断报告（放射学/病理学/实验室）、临床试验报告（ICH-E3、SAE、CSR）和患者文件（SOAP、H&P、出院小结）。全面支持模板、法规遵从性（HIPAA、FDA、ICH-GCP）和验证工具。

Constraint-based metabolic modeling (COBRA). FBA, FVA, gene knockouts, flux sampling, SBML models, for systems biology and metabolic engineering analysis. 基于约束的代谢模型（COBRA）。 FBA、FVA、基因敲除、通量采样、SBML 模型，用于系统生物学和代谢工程分析。

NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization. NGS 分析工具包。 BAM 到 bigWig 的转换、QC（相关性、PCA、指纹）、热图/概况（TSS、峰值），用于 ChIP-seq、RNA-seq、ATAC-seq 可视化。

Query the Cancer Dependency Map (DepMap) for cancer cell line gene dependency scores (CRISPR Chronos), drug sensitivity data, and gene effect profiles. Use for identifying cancer-specific vulnerabilities, synthetic lethal interactions, and validating oncology drug targets. 查询癌症依赖性图谱 (DepMap)，了解癌细胞系基因依赖性评分 (CRISPR Chronos)、药物敏感性数据和基因效应概况。用于识别癌症特定的脆弱性、合成致死相互作用以及验证肿瘤药物靶点。

Predict small-molecule binding poses with DiffDock-L (Corso et al. 2023/2024, github.com/gcorso/DiffDock) — blind diffusion docking that places a ligand into a protein pocket without a predefined search box and ranks the samples with a learned confidence model. Reach for this skill to dock a SMILES or SDF against a PDB, to generate ranked 3D poses for a small fragment library, or to get a starting pose for downstream rescoring. DiffDock predicts geometry, not affinity. 用 DiffDock-L 预测小分子结合位姿（Corso 等 2023/2024）——无需预定义搜索框的盲扩散对接，并用学习到的置信度模型对样本排序。用于将 SMILES/SDF 对接到 PDB、为小型片段库生成排序的 3D 位姿，或作为对接起点。

Use when working directly with the `esm` Python SDK, ESM3 or ESMC model IDs, Forge/Biohub inference clients, or ESMFold2 folding workflows. 用于蛋白质语言模型的综合工具包，包括 ESM3（跨序列、结构和功能的生成多模式蛋白质设计）和 ESM C（高效蛋白质嵌入和表示）。在处理蛋白质序列、结构或功能预测时使用此技能；设计新型蛋白质；生成蛋白质嵌入；进行反向折叠；或进行蛋白质工程任务。支持本地模型使用和基于云的 Forge API 以进行可扩展推理。

Biohub ESMFold2 / ESMFold2-Fast all-atom co-folding (Candido et al. 2026, github.com/Biohub/esm). Single-sequence and MSA modes; protein, DNA, RNA, ligand (CCD/SMILES), modified residues. FoldBench Ab-Ag 50-55%, PPI 70-77% DockQ-pass. Also covers the ESMC-{300M,600M,6B} protein language models from the same release: masked-LM logits, hidden states, mutation scoring, contact prediction, and the SAE interpretability head. MIT-licensed weights on HuggingFace org `biohub`. Use this skill when: (1) Predicting complex structures with single-sequence input, (2) Validating designed binders with ESMFold2-Fast, (3) Running ESMFold2 with MSA input, (4) Getting ESMC embeddings or per-residue mutation scores, (5) Choosing kernel backend and sampling-step settings for paper-faithful throughput. Biohub ESMFold2/ESMFold2-Fast 全原子共折叠（Candido 等 2026）。支持单序列与 MSA 模式，覆盖蛋白质/DNA/RNA/配体（CCD/SMILES）与修饰残基；同时包含 ESMC-{300M,600M,6B} 蛋白质语言模型：掩码 LM logits、隐状态、突变打分与接触预测。

Phylogenetic tree toolkit (ETE). Tree manipulation (Newick/NHX), evolutionary event detection, orthology/paralogy, NCBI taxonomy, visualization (PDF/SVG), for phylogenomics. 系统发育树工具包（ETE）。树操作 (Newick/NHX)、进化事件检测、直系同源/旁系同源、NCBI 分类、可视化 (PDF/SVG)，用于系统发育学。

Score, embed, and generate DNA sequences with Evo 2, a long-context genomic foundation model. Use this skill when: (1) Computing per-nucleotide or per-sequence likelihoods for variant effect scoring, (2) Embedding genomic windows for downstream classification, (3) Generating DNA conditioned on a prefix, (4) Scoring regulatory or coding regions across species. 用长上下文基因组基础模型 Evo 2 对 DNA 序列打分、嵌入与生成。用于计算逐核苷酸/整序列似然做变异效应打分、嵌入基因组窗口做下游分类、基于前缀生成 DNA，或跨物种为调控/编码区打分。

Embed proteins with Meta AI's ESM-2 (`fair-esm` package). Use this skill when: (1) Extracting per-residue or per-sequence embeddings for downstream ML, (2) Masked-LM likelihood / mutation effect scoring, (3) Contact prediction from a sequence. 用 Meta AI 的 ESM-2（fair-esm 包）嵌入蛋白质。用于为下游 ML 提取逐残基/整序列嵌入、做掩码 LM 似然/突变效应打分，或从序列预测接触图。

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. 解析 FCS（流式细胞术标准）文件 v2.0-3.1。将事件提取为 NumPy 数组，读取元数据/通道，转换为 CSV/DataFrame，用于流式细胞术数据预处理。

This skill should be used when working with genomic interval data (BED files) for machine learning tasks. Use for training region embeddings (Region2Vec, BEDspace), single-cell ATAC-seq analysis (scEmbed), building consensus peaks (universes), or any ML-based analysis of genomic regions. Applies to BED file collections, scATAC-seq data, chromatin accessibility datasets, and region-based genomic feature learning. 在处理机器学习任务的基因组区间数据（BED 文件）时应使用此技能。用于训练区域嵌入（Region2Vec、BEDspace）、单细胞 ATAC-seq 分析 (scEmbed)、构建共识峰（宇宙）或任何基于 ML 的基因组区域分析。适用于 BED 文件集合、scATAC-seq 数据、染色质可及性数据集和基于区域的基因组特征学习。

Fast CLI/Python queries to 20+ bioinformatics databases. Use for quick lookups: gene info, BLAST/BLAT, viral sequence downloads, AlphaFold structures, enrichment analysis, OpenTargets, COSMIC, CELLxGENE, and 8cube mouse specificity/expression data. Best for interactive exploration and simple queries. For batch processing or advanced BLAST use biopython; for multi-database Python workflows use bioservices. 对 20 多个生物信息学数据库进行快速 CLI/Python 查询。用于快速查找：基因信息、BLAST 搜索、AlphaFold 结构、富集分析。最适合交互式探索、简单查询。对于批处理或高级 BLAST，请使用 biopython；对于多数据库Python工作流程，使用bioservices。

Analyze and engineer protein glycosylation. Scan sequences for N-glycosylation sequons (N-X-S/T), predict O-glycosylation hotspots, and access curated glycoengineering tools (NetOGlyc, GlycoShield, GlycoWorkbench). For glycoprotein engineering, therapeutic antibody optimization, and vaccine design. 分析和设计蛋白质糖基化。扫描 N-糖基化序列 (N-X-S/T) 的序列，预测 O-糖基化热点，并访问精选的糖工程工具（NetOGlyc、GlycoShield、GlycoWorkbench）。用于糖蛋白工程、治疗性抗体优化和疫苗设计。

High-performance toolkit for genomic interval analysis in Rust with Python bindings. Use when working with genomic regions, BED files, coverage tracks, overlap detection, tokenization for ML models, or fragment analysis in computational genomics and machine learning applications. 使用 Rust 与 Python 绑定进行基因组区间分析的高性能工具包。在处理基因组区域、BED 文件、覆盖轨迹、重叠检测、ML 模型标记化或计算基因组学和机器学习应用程序中的片段分析时使用。

Lightweight WSI tile extraction and preprocessing. Use for basic slide processing, tissue detection, tile extraction, and stain normalization for H&E images. Best for simple pipelines, dataset preparation, and quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml. 轻量级 WSI 切片提取和预处理。用于 H&E 图像的基本载玻片处理组织检测、平铺提取、染色归一化。最适合简单的管道、数据集准备、基于图块的快速分析。对于高级空间蛋白质组学、多重成像或深度学习管道，请使用 pathml。

Query and download public cancer imaging data from NCI Imaging Data Commons using idc-index. Use for accessing large-scale radiology (CT, MR, PET) and pathology datasets for AI training or research. No authentication required. Query by metadata, visualize in browser, check licenses. 使用 idc-index 从 NCI Imaging Data Commons 查询和下载公共癌症成像数据。用于访问大规模放射学（CT、MR、PET）和病理学数据集以进行 AI 培训或研究。无需身份验证。按元数据查询、在浏览器中可视化、检查许可证。

Comprehensive toolkit for preparing ISO 13485 certification documentation for medical device Quality Management Systems. Use when users need help with ISO 13485 QMS documentation, including (1) conducting gap analysis of existing documentation, (2) creating Quality Manuals, (3) developing required procedures and work instructions, (4) preparing Medical Device Files, (5) understanding ISO 13485 requirements, or (6) identifying missing documentation for medical device certification. Also use when users mention medical device regulations, QMS certification, FDA QMSR, EU MDR, or need help with quality system documentation. 用于为医疗器械质量管理体系准备 ISO 13485 认证文档的综合工具包。当用户需要 ISO 13485 QMS 文档方面的帮助时使用，包括 (1) 对现有文档进行差距分析，(2) 创建质量手册，(3) 制定所需的程序和工作说明，(4) 准备医疗器械文件，(5) 了解 ISO 13485 要求，或 (6) 识别医疗器械认证缺失的文档。当用户提及医疗器械法规、QMS 认证、FDA QMSR、EU MDR 或需要质量体系文档帮助时也可使用。

Use when working with LaminDB, the open-source lineage-native lakehouse for biological datasets and models. Covers setup, artifact registration, query/search, lineage tracking, validation, ontology-backed annotation with Bionty, collections, branches, storage, and workflow integrations. 在使用 LaminDB 时应该使用此技能，LaminDB 是一种生物学开源数据框架，使数据可查询、可追踪、可再现且公平。在管理生物数据集（scRNA-seq、空间、流式细胞术等）、跟踪计算工作流程、使用生物本体整理和验证数据、构建数据湖房或确保生物研究中的数据沿袭和可重复性时使用。涵盖数据管理、注释、本体（基因、细胞类型、疾病、组织）、模式验证、与工作流程管理器（Nextflow、Snakemake）和 MLOps 平台（W&B、MLflow）的集成以及部署策略。

Inverse-fold a backbone with ligand, nucleic-acid, and metal context using LigandMPNN (Dauparas et al. 2023, github.com/dauparas/LigandMPNN). Reach for this skill to redesign the residues lining a binding pocket around a bound small molecule or cofactor, to design metal-coordinating sites where the geometry must be respected, or to get threaded designed-sequence PDBs out of any MPNN run. 用 LigandMPNN 在配体、核酸与金属环境下对骨架做逆折叠（Dauparas 等 2023）。用于围绕结合的小分子或辅因子重设计口袋衬里残基、设计需尊重几何的金属配位位点，或从任意 MPNN 运行输出穿线后的设计序列 PDB。

Comprehensive biosignal processing toolkit for analyzing physiological data including ECG, EEG, EDA, RSP, PPG, EMG, and EOG signals. Use this skill when processing cardiovascular signals, brain activity, electrodermal responses, respiratory patterns, muscle activity, or eye movements. Applicable for heart rate variability analysis, event-related potentials, complexity measures, autonomic nervous system assessment, psychophysiology research, and multi-modal physiological signal integration. 用于分析生理数据的综合生物信号处理工具包，包括 ECG、EEG、EDA、RSP、PPG、EMG 和 EOG 信号。在处理心血管信号、大脑活动、皮肤电反应、呼吸模式、肌肉活动或眼球运动时使用此技能。适用于心率变异性分析、事件相关电位、复杂性测量、自主神经系统评估、心理生理学研究和多模态生理信号集成。

Analyze Neuropixels extracellular recordings end-to-end with SpikeInterface. Covers loading SpikeGLX/Open Ephys/NWB data, preprocessing, drift/motion correction, Kilosort4 (and CPU) spike sorting, quality metrics, and unit curation (threshold-based, model-based UnitRefine, and AI-assisted visual review). Use when working with Neuropixels 1.0/2.0 recordings, spike sorting, or extracellular electrophysiology analysis. Neuropixels 神经记录分析。加载 SpikeGLX/OpenEphys 数据、预处理、运动校正、Kilosort4 尖峰排序、质量指标、Allen/IBL 管理、AI 辅助视觉分析，适用于 Neuropixels 1.0/2.0 细胞外电生理学。当处理神经记录、尖峰排序、细胞外电生理学时，或者当用户提到 Neuropixels、SpikeGLX、Open Ephys、Kilosort、质量指标或单位管理时使用。

Microscopy data management platform. Access images via Python, retrieve datasets, analyze pixels, manage ROIs/annotations, batch processing, for high-content screening and microscopy workflows. 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 ROI/注释、批处理，以实现高内容筛选和显微镜工作流程。

Query the 1000 Genomes Project dataset (3,202 whole-genome-sequenced individuals, GRCh38) at the level of individual participants. Use when a question is about individuals or variants in the 1000 Genomes Project cohort: which individuals carry variants matching specific criteria in a gene or region, which individuals are homozygous-reference at a position, which variants exist in the dataset or carried by specified individuals in a gene or region, the relatedness between two specified individuals. Variants are returned with 1000 Genomes allele frequencies (AF), gnomAD v4.1 exome and genome AF, AlphaMissense score, and HGVSp annotations. 千人基因组计划（1000 Genomes，GRCh38，3202 例全基因组测序个体）个体级查询。用于查找携带特定基因/区域变异的个体、位点纯合参考个体、数据集或指定个体的变异，以及个体间亲缘关系；变异附带千人基因组与 gnomAD v4.1 等位基因频率、AlphaMissense 评分及 HGVSp 注释。

Structure prediction using OpenFold3, an open-weights PyTorch reproduction of AlphaFold3 from the AlQuraishi Lab. Use this skill when predicting protein/nucleic-acid/ligand complex structures with an Apache-2.0-licensed AF3 reimplementation. 用 OpenFold3 做结构预测——AlQuraishi 实验室对 AlphaFold3 的开源权重 PyTorch 复现。当需要用 Apache-2.0 许可的 AF3 复现来预测蛋白质/核酸/配体复合物结构时使用。

Operator toolkit for nf-core/pacsomatic matched tumor-normal workflows from BAM inputs. Use this skill when the user needs to validate run inputs, generate pacsomatic-compliant samplesheets, prepare reproducible Nextflow launch artifacts, run locally or submit to schedulers (LSF/Slurm/PBS/SGE), and triage execution failures. Triggers on requests to run pacsomatic, prepare launch commands/scripts, perform dry-run checks, or troubleshoot pipeline startup and scheduler submission errors. nf-core/pacsomatic 匹配肿瘤-正常样本 BAM 输入工作流的操作工具包。用于验证运行输入、生成符合 pacsomatic 的 samplesheet、准备可复现的 Nextflow 启动 artifacts、本地运行或提交到 LSF/Slurm/PBS/SGE，并排查执行失败。

Full-featured computational pathology toolkit. Use for advanced WSI analysis including multiplexed immunofluorescence (CODEX, Vectra), nucleus segmentation, tissue graph construction, and ML model training on pathology data. Supports 160+ slide formats. For simple tile extraction from H&E slides, histolab may be simpler. 全功能的计算病理学工具包。用于高级 WSI 分析，包括多重免疫荧光（CODEX、Vectra）、细胞核分割、组织图构建和病理数据的 ML 模型训练。支持 160 多种幻灯片格式。对于从 H&E 载玻片中进行简单的切片提取，histolab 可能更简单。

Run pathway and gene-set enrichment analysis on gene lists or ranked gene data, then interpret the results. Use whenever the user has a set of genes (differentially expressed genes from PyDESeq2/Scanpy, CRISPR-screen hits, cluster marker genes, proteomics hits) and wants to know which biological pathways, GO terms, or gene sets are over-represented or enriched. Covers over-representation analysis (ORA / Enrichr / Fisher / hypergeometric), ranked Gene Set Enrichment Analysis (GSEA / preranked), single-sample scoring (ssGSEA/GSVA), and functional profiling via gseapy, g:Profiler, Enrichr libraries, MSigDB, GO, KEGG, Reactome, and WikiPathways — plus gene-ID mapping, choosing the right background universe, multiple-testing correction, redundancy reduction, dotplots/enrichment maps, and publication-ready tables. Use this for "pathway analysis", "enrichment analysis", "GO enrichment", "KEGG/Reactome pathways", "GSEA", "over-representation", "functional annotation", or "what pathways are my genes in". 对基因列表或排序基因数据运行通路和基因集富集分析并解释结果。用于差异表达基因、CRISPR 筛选命中、cluster marker、蛋白质组学命中等，覆盖 ORA/Enrichr/Fisher/超几何检验、GSEA/preranked、ssGSEA/GSVA、gseapy、g:Profiler、MSigDB、GO、KEGG、Reactome、WikiPathways，以及基因 ID 映射、背景集选择、多重检验校正和发表级表格/图。

Build and analyze phylogenetic trees using MAFFT (multiple alignment), IQ-TREE 2 (maximum likelihood), and FastTree (fast NJ/ML). Visualize with ETE3 or FigTree. For evolutionary analysis, microbial genomics, viral phylodynamics, protein family analysis, and molecular clock studies. 使用 MAFFT（多重比对）、IQ-TREE 2（最大似然）和 FastTree（快速 NJ/ML）构建和分析系统发育树。使用 ETE3 或 FigTree 进行可视化。用于进化分析、微生物基因组学、病毒系统动力学、蛋白质家族分析和分子钟研究。

Query the Precision Medicine Knowledge Graph (PrimeKG) for multiscale biological data including genes, drugs, diseases, phenotypes, and more. 查询精准医学知识图（PrimeKG）以获取多尺度生物数据，包括基因、药物、疾病、表型等。

Inverse-fold a protein backbone (PDB structure) into amino-acid sequence with ProteinMPNN (Dauparas et al. 2022, github.com/dauparas/ProteinMPNN). Reach for this skill to run sequence design on RFdiffusion backbones, to redesign one chain of a PDB while holding interface residues fixed, or to generate a temperature-swept set of sequences for downstream folding. 用 ProteinMPNN 将蛋白质骨架（PDB 结构）逆折叠为氨基酸序列（Dauparas 等 2022）。用于对 RFdiffusion 骨架做序列设计、在固定界面残基的同时重设计 PDB 的某条链，或生成温度扫描的一组序列供下游折叠。

Differential gene expression analysis for bulk RNA-seq with PyDESeq2, including formulaic designs, Wald tests, FDR correction, LFC shrinkage, and result visualization. 差异基因表达分析（Python DESeq2）。从批量 RNA-seq 计数、Wald 检验、FDR 校正、火山/MA 图中识别 DE 基因，以进行 RNA-seq 分析。

Python library for working with DICOM (Digital Imaging and Communications in Medicine) files. Use this skill when reading, writing, or modifying medical imaging data in DICOM format, extracting pixel data from medical images (CT, MRI, X-ray, ultrasound), anonymizing DICOM files, working with DICOM metadata and tags, converting DICOM images to other formats, handling compressed DICOM data, or processing medical imaging datasets. Applies to tasks involving medical image analysis, PACS systems, radiology workflows, and healthcare imaging applications. 用于处理 DICOM（医学数字成像和通信）文件的 Python 库。在读取、写入或修改 DICOM 格式的医学成像数据、从医学图像（CT、MRI、X 射线、超声）中提取像素数据、匿名 DICOM 文件、使用 DICOM 元数据和标签、将 DICOM 图像转换为其他格式、处理压缩的 DICOM 数据或处理医学成像数据集时，请使用此技能。适用于涉及医学图像分析、PACS 系统、放射学工作流程和医疗保健成像应用程序的任务。

Build clinical/healthcare deep-learning pipelines with PyHealth — loading EHR/signal/imaging datasets (MIMIC-III/IV, eICU, OMOP, SleepEDF, ChestXray14, EHRShot), defining tasks (mortality, readmission, length-of-stay, drug recommendation, sleep staging, ICD coding, EEG events), instantiating models (Transformer, RETAIN, GAMENet, SafeDrug, MICRON, StageNet, AdaCare, CNN/RNN/MLP), training with the PyHealth Trainer, computing clinical metrics, and using medical code utilities (ICD/ATC/NDC/RxNorm lookup and cross-mapping). Use this skill whenever the user mentions PyHealth, MIMIC, eICU, OMOP, EHR modeling, clinical prediction, drug recommendation, sleep staging, medical code mapping, ICD/ATC codes, or any healthcare ML pipeline that fits the dataset → task → model → trainer → metrics pattern, even if "PyHealth" isn't named explicitly. 使用 PyHealth 构建临床/医疗保健深度学习管道 — 加载 EHR/信号/成像数据集（MIMIC-III/IV、eICU、OMOP、SleepEDF、ChestXray14、EHRShot）、定义任务（死亡率、再入院、住院时间、药物推荐、睡眠分期、ICD 编码、EEG 事件）、实例化模型（Transformer、RETAIN、GAMENet、SafeDrug、MICRON、 StageNet、AdaCare、CNN/RNN/MLP）、使用 PyHealth Trainer 进行训练、计算临床指标以及使用医疗代码实用程序（ICD/ATC/NDC/RxNorm 查找和交叉映射）。每当用户提及 PyHealth、MIMIC、eICU、OMOP、EHR 建模、临床预测、药物推荐、睡眠分期、医疗代码映射、ICD/ATC 代码或任何适合数据集 → 任务 → 模型 → 训练器 → 指标模式的医疗保健 ML 管道时，即可使用此技能，即使“PyHealth”未明确命名。

Complete mass spectrometry analysis platform. Use for proteomics and metabolomics workflows—feature detection, peptide/protein identification, label-free and isobaric quantification, adduct/accurate-mass annotation, and complex LC-MS/MS pipelines. Supports extensive file formats and algorithms. For simple spectral comparison and small-molecule library matching use matchms. 完整的质谱分析平台。用于蛋白质组学工作流程，包括检测、肽鉴定、蛋白质定量和复杂的 LC-MS/MS 流程。支持广泛的文件格式和算法。最适合蛋白质组学、全面的 MS 数据处理。对于简单的光谱比较和代谢物 ID，请使用 matchms。

Genomic file toolkit. Read/write SAM/BAM/CRAM alignments, VCF/BCF variants, FASTA/FASTQ sequences, extract regions, calculate coverage, for NGS data processing pipelines. 基因组文件工具包。读/写 SAM/BAM/CRAM 比对、VCF/BCF 变体、FASTA/FASTQ 序列、提取区域、计算覆盖率，用于 NGS 数据处理流程。

Therapeutics Data Commons. AI-ready drug discovery datasets (ADME, toxicity, DTI), benchmarks, scaffold splits, molecular oracles, for therapeutic ML and pharmacological prediction. 治疗数据共享。 AI 就绪药物发现数据集（ADME、毒性、DTI）、基准、支架分割、分子预言，用于治疗性 ML 和药理学预测。

Standard single-cell RNA-seq analysis pipeline. Use for QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, visualization, and converting R-friendly single-cell formats such as Seurat or SingleCellExperiment RDS files into h5ad for Scanpy. Best for exploratory scRNA-seq analysis with established workflows. For deep learning models use scvi-tools; for data format questions use anndata. 标准单细胞 RNA-seq 分析流程。用于 QC、归一化、降维 (PCA/UMAP/t-SNE)、聚类、差异表达和可视化。最适合使用已建立的工作流程进行探索性 scRNA-seq 分析。对于深度学习模型，请使用 scvi-tools；对于数据格式问题，请使用 anndata。

Embed and annotate single-cell expression data with scGPT, a foundation model for single-cell biology. Use this skill when: (1) Producing cell embeddings from an AnnData for clustering/integration, (2) Zero-shot or fine-tuned cell-type annotation, (3) Gene-level representation for perturbation/GRN tasks. 用单细胞生物学基础模型 scGPT 嵌入并注释单细胞表达数据。用于从 AnnData 生成细胞嵌入做聚类/整合、零样本或微调的细胞类型注释，或为扰动/GRN 任务生成基因级表征。

Biological data toolkit. Sequence analysis, alignments, phylogenetic trees, diversity metrics (alpha/beta, UniFrac), ordination (PCoA), PERMANOVA, FASTA/Newick I/O, for microbiome analysis. 生物数据工具包。序列分析、比对、系统发育树、多样性指标（α/β、UniFrac）、排序（PCoA）、PERMANOVA、FASTA/Newick I/O，用于微生物组分析。

Comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival. Use this skill when working with censored survival data, performing time-to-event analysis, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating survival predictions with concordance index or Brier score, handling competing risks, or implementing any survival analysis workflow with the scikit-survival library. 使用 scikit-survival 在 Python 中进行生存分析和事件时间建模的综合工具包。在处理经过审查的生存数据、执行事件时间分析、拟合 Cox 模型、随机生存森林、梯度提升模型或生存 SVM、使用一致性指数或 Brier 评分评估生存预测、处理竞争风险或使用 scikit-survival 库实施任何生存分析工作流程时，可以使用此技能。

RNA velocity analysis with scVelo. Estimate cell state transitions from unspliced/spliced mRNA dynamics, infer trajectory directions, compute latent time, and identify driver genes in single-cell RNA-seq data. Complements Scanpy/scVI-tools for trajectory inference. 使用 scVelo 进行 RNA 速度分析。根据未剪接/剪接 mRNA 动力学估计细胞状态转变，推断轨迹方向，计算潜伏时间，并识别单细胞 RNA-seq 数据中的驱动基因。补充了用于轨迹推断的 Scanpy/scVI 工具。

Deep generative models for single-cell omics. Use when you need probabilistic batch correction (scVI), transfer learning, differential expression with uncertainty, or multi-modal integration (TOTALVI, MultiVI). Best for advanced modeling, batch effects, multimodal data. For standard analysis pipelines use scanpy. 单细胞组学的深度生成模型。当您需要概率批量校正 (scVI)、迁移学习、不确定性差异表达或多模态积分（TOTALVI、MultiVI）时使用。最适合高级建模、批量效果、多模式数据。对于标准分析管道，请使用 scanpy。

Probabilistic single-cell RNA-seq with scvi-tools — scVI for a batch-corrected latent space, scANVI for semi-supervised label transfer, and Bayesian differential expression. Reach for this skill to integrate scRNA-seq batches, embed cells for clustering, transfer annotations from a reference onto a query, or score differentially expressed genes per cluster. For spatial deconvolution / mapping use the cell2location, DestVI, or Tangram methods instead. 用 scvi-tools 做概率式单细胞 RNA-seq——scVI 生成批次校正的隐空间、scANVI 做半监督标签迁移、贝叶斯差异表达。用于整合 scRNA-seq 批次、嵌入细胞聚类、将参考注释迁移到查询集，或按簇为差异表达基因打分。

Inverse-fold a backbone with SolubleMPNN — ProteinMPNN retrained on a soluble-PDB subset (Dauparas et al. 2022) — for sequences biased toward cytosolic expression and reduced aggregation. Reach for this skill when designs from vanilla ProteinMPNN are aggregating or going to inclusion bodies, when redesigning a membrane-adjacent fold for soluble expression, or when an E. coli expression screen is the next step. 用 SolubleMPNN 对骨架逆折叠——在可溶 PDB 子集上重训练的 ProteinMPNN（Dauparas 等 2022），使序列偏向胞质表达并减少聚集。当 vanilla ProteinMPNN 的设计发生聚集或进入包涵体、需为可溶表达重设计近膜折叠，或做 E. coli 表达筛选时使用。

Access a collection of open-source molecular design and structural biology tools on the Tamarind Bio platform, via its REST API or MCP server — no local GPUs required. Tamarind bundles popular open-source models for structure prediction (AlphaFold, Boltz, Chai, ESMFold), protein, binder, and de novo design (RFdiffusion, ProteinMPNN, BoltzGen), antibody and nanobody design and developability, protein-ligand docking (DiffDock, Autodock Vina), binding-affinity prediction, MSA generation, and molecular dynamics. Use when the user mentions Tamarind or tamarind.bio, wants to run any of these open-source tools in the cloud, references app.tamarind.bio/api or the x-api-key header, or needs to submit batches of sequences for structural or biophysical characterization. Tamarind Bio 平台上的开源分子设计与结构生物学工具集，通过 REST API 或 MCP 服务器调用，无需本地 GPU。集成 AlphaFold、Boltz、Chai、ESMFold 等结构预测，以及 RFdiffusion 等蛋白、结合体与从头设计模型。

Efficient storage and retrieval of genomic variant data using TileDB. Scalable VCF/BCF ingestion, incremental sample addition, compressed storage, parallel queries, and export capabilities for population genomics. 使用 TileDB 高效存储和检索基因组变异数据。可扩展的 VCF/BCF 摄取、增量样本添加、压缩存储、并行查询和群体基因组学导出功能。

Generate concise (3-4 page), focused medical treatment plans in LaTeX/PDF format for all clinical specialties. Supports general medical treatment, rehabilitation therapy, mental health care, chronic disease management, perioperative care, and pain management. Includes SMART goal frameworks, evidence-based interventions with minimal text citations, regulatory compliance (HIPAA), and professional formatting. Prioritizes brevity and clinical actionability. 为所有临床专业生成 LaTeX/PDF 格式的简明（3-4 页）、重点突出的医疗计划。支持一般医疗、康复治疗、心理保健、慢性病管理、围手术期护理和疼痛管理。包括 SMART 目标框架、以最少文本引用的循证干预措施、法规遵从性 (HIPAA) 和专业格式。优先考虑简洁性和临床可操作性。

Core Python library for astronomy and astrophysics workflows that need Astropy APIs, including units/quantities, coordinates, FITS I/O, tables, time systems, WCS, and cosmology. Use when implementing or debugging astronomical data analysis code with Astropy. 用于天文学和天体物理学的综合 Python 库。在处理天文数据（包括天体坐标、物理单位、FITS 文件、宇宙学计算、时间系统、表格、世界坐标系 (WCS) 和天文数据分析）时，应使用此技能。当任务涉及坐标变换、单位转换、FITS 文件操作、宇宙距离计算、时间尺度转换或天文数据处理时使用。

Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for Google hardware, noise modeling, and low-level circuit design. For IBM hardware use qiskit; for quantum ML with autodiff use pennylane; for physics simulations use qutip. 谷歌量子计算框架。在针对 Google Quantum AI 硬件、设计噪声感知电路或运行量子表征实验时使用。最适合 Google 硬件、噪声建模和低级电路设计。对于 IBM 硬件，请使用 qiskit；对于具有自动差分功能的量子 ML，请使用 pennylane；对于物理模拟，请使用 qutip。

Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use rdkit directly. RDKit 的 Pythonic 包装器，具有简化的界面和合理的默认值。标准药物发现的首选，包括 SMILES 解析、标准化、描述符、指纹、聚类、3D 构象异构体、并行处理。返回本机 rdkit.Chem.Mol 对象。对于高级控制或自定义参数，直接使用rdkit。

Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc. 具有多种特征器和预构建数据集的分子机器学习。当您需要广泛的特征化选项和 MoleculeNet 基准时，可使用传统 ML 或 GNN 进行属性预测（ADMET、毒性）。最适合使用预先训练的模型、不同的分子表示进行快速实验。对于图优先的 PyTorch 工作流程，请使用 torchdrug；对于基准数据集，请使用 pytdc。

DiffDock and DiffDock-L molecular docking. Use for protein-small-molecule pose prediction from PDB or sequence plus SMILES/SDF/MOL2, batch docking, virtual screening, and pose-confidence interpretation. Not for binding affinity prediction. 基于扩散的分子对接。通过 PDB/SMILES、置信度评分、虚拟筛选预测蛋白质-配体结合姿势，以进行基于结构的药物设计。不适用于亲和力预测。

Framework for computational fluid dynamics simulations using Python. Use when running fluid dynamics simulations including Navier-Stokes equations (2D/3D), shallow water equations, stratified flows, or when analyzing turbulence, vortex dynamics, or geophysical flows. Provides pseudospectral methods with FFT, HPC support, and comprehensive output analysis. 使用 Python 进行计算流体动力学模拟的框架。在运行流体动力学模拟（包括纳维-斯托克斯方程 (2D/3D)、浅水方程、分层流）或分析湍流、涡动力学或地球物理流时使用。提供具有 FFT、HPC 支持和全面输出分析的伪谱方法。

Spectral similarity and compound identification for metabolomics. Use for comparing mass spectra, computing similarity scores (cosine, modified cosine), and identifying unknown compounds from spectral libraries. Best for metabolite identification, spectral matching, library searching. For full LC-MS/MS proteomics pipelines use pyopenms. 代谢组学的光谱相似性和化合物鉴定。用于比较质谱、计算相似性分数（余弦、修正余弦）以及从谱库中识别未知化合物。最适合代谢物鉴定、光谱匹配、库搜索。对于完整的 LC-MS/MS 蛋白质组学管道，请使用 pyopenms。

MATLAB and GNU Octave numerical computing for matrix operations, data analysis, visualization, and scientific computing. Use when writing MATLAB/Octave scripts for linear algebra, signal processing, image processing, differential equations, optimization, statistics, or creating scientific visualizations. Also use when the user needs help with MATLAB syntax, functions, or wants to convert between MATLAB and Python code. Scripts can be executed with MATLAB or the open-source GNU Octave interpreter. MATLAB 和 GNU Octave 数值计算，用于矩阵运算、数据分析、可视化和科学计算。在编写用于线性代数、信号处理、图像处理、微分方程、优化、统计或创建科学可视化的 MATLAB/Octave 脚本时使用。当用户需要有关 MATLAB 语法、函数的帮助或想要在 MATLAB 和 Python 代码之间进行转换时，也可使用。脚本可以使用 MATLAB 或开源 GNU Octave 解释器执行。

Medicinal chemistry filters for compound triage. Apply drug-likeness rules (Lipinski, Veber, CNS), structural alert catalogs (PAINS, NIBR, ChEMBL), complexity metrics, and the medchem query language for library filtering. 药物化学过滤器。应用药物相似性规则（Lipinski、Veber）、PAINS 过滤器、结构警报、复杂性指标，以进行化合物优先级排序和库过滤。

Run and analyze molecular dynamics simulations with OpenMM and MDAnalysis. Set up protein/small molecule systems, define force fields, run energy minimization and production MD, analyze trajectories (RMSD, RMSF, contact maps, free energy surfaces). For structural biology, drug binding, and biophysics. 使用 OpenMM 和 MDAnalysis 运行和分析分子动力学模拟。设置蛋白质/小分子系统，定义力场，运行能量最小化和生产MD，分析轨迹（RMSD、RMSF、接触图、自由能表面）。用于结构生物学、药物结合和生物物理学。

Molecular featurization for ML (100+ featurizers). ECFP, MACCS, descriptors, pretrained models (ChemBERTa), convert SMILES to features, for QSAR and molecular ML. ML 的分子特征化（100 多个特征化器）。 ECFP、MACCS、描述符、预训练模型 (ChemBERTa)、将 SMILES 转换为特征，用于 QSAR 和分子 ML。

Hardware-agnostic quantum ML framework with automatic differentiation. Use when training quantum circuits via gradients, building hybrid quantum-classical models, or needing device portability across IBM/Google/Rigetti/IonQ. Best for variational algorithms (VQE, QAOA), quantum neural networks, and integration with PyTorch or JAX. For hardware-specific optimizations use qiskit (IBM) or cirq (Google); for open quantum systems use qutip. 具有自动微分功能的与硬件无关的量子机器学习框架。当通过梯度训练量子电路、构建混合量子经典模型或需要跨 IBM/Google/Rigetti/IonQ 的设备可移植性时使用。最适合变分算法（VQE、QAOA）、量子神经网络以及与 PyTorch/JAX/TensorFlow 的集成。对于特定于硬件的优化，请使用 qiskit (IBM) 或 cirq (Google)；对于开放量子系统，请使用 qutip。

Materials science toolkit. Crystal structures (CIF, POSCAR), phase diagrams, band structure, DOS, Materials Project integration, format conversion, for computational materials science. 材料科学工具包。晶体结构（CIF、POSCAR）、相图、能带结构、DOS、材料项目集成、格式转换，用于计算材料科学。

IBM quantum computing framework. Use when targeting IBM Quantum hardware, working with Qiskit Runtime for production workloads, or needing IBM optimization tools. Best for IBM hardware execution, quantum error mitigation, and enterprise quantum computing. For Google hardware use cirq; for gradient-based quantum ML use pennylane; for open quantum system simulations use qutip. IBM 量子计算框架。当针对 IBM Quantum 硬件、使用 Qiskit Runtime 处理生产工作负载或需要 IBM 优化工具时使用。最适合 IBM 硬件执行、量子错误缓解和企业量子计算。对于 Google 硬件，请使用 cirq；对于基于梯度的量子机器学习，使用 pennylane；对于开放量子系统模拟，请使用 qutip。

Quantum physics simulation library for open quantum systems. Use when studying master equations, Lindblad dynamics, decoherence, quantum optics, or cavity QED. Best for physics research, open system dynamics, and educational simulations. NOT for circuit-based quantum computing—use qiskit, cirq, or pennylane for quantum algorithms and hardware execution. 开放量子系统的量子物理模拟库。在研究主方程、Lindblad 动力学、退相干、量子光学或腔 QED 时使用。最适合物理研究、开放系统动力学和教育模拟。不适用于基于电路的量子计算 - 使用 qiskit、cirq 或 pennylane 进行量子算法和硬件执行。

Cheminformatics toolkit for fine-grained molecular control. SMILES/SDF parsing, descriptors (MW, LogP, TPSA), fingerprints, substructure search, 2D/3D generation, similarity, reactions. For standard workflows with simpler interface, use datamol (wrapper around RDKit). Use rdkit for advanced control, custom sanitization, specialized algorithms. 用于细粒度分子控制的化学信息学工具包。 SMILES/SDF 解析、描述符（MW、LogP、TPSA）、指纹、子结构搜索、2D/3D 生成、相似性、反应。对于具有更简单界面的标准工作流程，请使用 datamol（RDKit 的包装器）。使用 rdkit 进行高级控制、自定义清理和专用算法。

Rowan is a cloud-native molecular modeling and medicinal-chemistry workflow platform with a Python API. Use for pKa and macropKa prediction, conformer and tautomer ensembles, docking and analogue docking, protein-ligand cofolding, MSA generation, molecular dynamics, permeability, descriptor workflows, and related small-molecule or protein modeling tasks. Ideal for programmatic batch screening, multi-step chemistry pipelines, and workflows that would otherwise require maintaining local HPC/GPU infrastructure. Rowan 是一个具有 Python API 的云原生分子建模和药物化学工作流程平台。用于 pKa 和 macropKa 预测、构象异构体和互变异构体集成、对接和类似物对接、蛋白质-配体共折叠、MSA 生成、分子动力学、渗透性、描述符工作流程以及相关的小分子或蛋白质建模任务。非常适合程序化批量筛选、多步骤化学管道以及需要维护本地 HPC/GPU 基础设施的工作流程。

Use when you need exact symbolic math in Python — algebra, calculus, equation solving, symbolic linear algebra, or code generation via lambdify/LaTeX. Prefer NumPy or SciPy when floating-point approximations are sufficient. 在 Python 中处理符号数学时可以使用此技能。该技能应用于符号计算任务，包括代数求解方程、执行微积分运算（导数、积分、极限）、操作代数表达式、符号处理矩阵、物理计算、数论问题、几何计算以及从数学表达式生成可执行代码。当用户需要精确的符号结果而不是数值近似值时，或者在使用包含变量和参数的数学公式时，可以应用此技能。

PyTorch-native graph neural networks for molecules and proteins. Use when building custom GNN architectures for drug discovery, protein modeling, or knowledge graph reasoning. Best for custom model development, protein property prediction, retrosynthesis. For pre-trained models and diverse featurizers use deepchem; for benchmark datasets use pytdc. PyTorch-分子和蛋白质的原生图神经网络。在构建用于药物发现、蛋白质建模或知识图推理的自定义 GNN 架构时使用。最适合定制模型开发、蛋白质特性预测、逆合成。对于预先训练的模型和不同的特征器，使用 deepchem；对于基准数据集，请使用 pytdc。

Comprehensive geospatial science skill covering remote sensing, GIS, spatial analysis, machine learning for earth observation, and 30+ scientific domains. Supports satellite imagery processing (Sentinel, Landsat, MODIS, SAR, hyperspectral), vector and raster data operations, spatial statistics, point cloud processing, network analysis, cloud-native workflows (STAC, COG, Planetary Computer), and 8 programming languages (Python, R, Julia, JavaScript, C++, Java, Go, Rust) with 500+ code examples. Use for remote sensing workflows, GIS analysis, spatial ML, Earth observation data processing, terrain analysis, hydrological modeling, marine spatial analysis, atmospheric science, and any geospatial computation task. 全面的地理空间科学技能，涵盖遥感、GIS、空间分析、地球观测机器学习和 30 多个科学领域。支持卫星图像处理（Sentinel、Landsat、MODIS、SAR、高光谱）、矢量和栅格数据操作、空间统计、点云处理、网络分析、云原生工作流程（STAC、COG、行星计算机）和 8 种编程语言（Python、R、Julia、JavaScript、C++、Java、Go、Rust）以及 500 多个代码示例。用于遥感工作流程、GIS 分析、空间 ML、地球观测数据处理、地形分析、水文建模、海洋空间分析、大气科学和任何地理空间计算任务。

Python library for working with geospatial vector data including shapefiles, GeoJSON, and GeoPackage files. Use when working with geographic data for spatial analysis, geometric operations, coordinate transformations, spatial joins, overlay operations, choropleth mapping, or any task involving reading/writing/analyzing vector geographic data. Supports PostGIS databases, interactive maps, and integration with matplotlib/folium/cartopy. Use for tasks like buffer analysis, spatial joins between datasets, dissolving boundaries, clipping data, calculating areas/distances, reprojecting coordinate systems, creating maps, or converting between spatial file formats. 用于处理地理空间矢量数据（包括 shapefile、GeoJSON 和 GeoPackage 文件）的 Python 库。在处理地理数据进行空间分析、几何运算、坐标变换、空间连接、叠加操作、分区统计图或涉及读取/写入/分析矢量地理数据的任何任务时使用。支持 PostGIS 数据库、交互式地图以及与 matplotlib/folium/cartopy 的集成。用于缓冲区分析、数据集之间的空间连接、溶解边界、剪切数据、计算面积/距离、重新投影坐标系、创建地图或在空间文件格式之间进行转换等任务。

Query the U.S. Treasury Fiscal Data REST API for federal financial data. No API key required. Use for national debt (Debt to the Penny), Daily Treasury Statements, Monthly Treasury Statements, Treasury securities auctions, interest rates, foreign exchange rates, savings bonds, or U.S. government revenue and spending statistics. 查询美国财政部财政数据 API 以获取联邦金融数据，包括国债、政府支出、收入、利率、汇率和储蓄债券。无需 API 密钥即可访问 54 个数据集和 182 个数据表。在处理美国联邦财政数据、国债跟踪（便士债务）、每日国债报表、月度国债报表、国债拍卖、国债利率、外汇汇率、储蓄债券或任何美国政府财务统计数据时使用。

Benchling Python SDK and REST API integration for registry entities, inventory, ELN entries, workflows, Benchling Apps, and Data Warehouse queries. Use when automating lab data with benchling-sdk or the v2 API. Benchling研发平台整合。通过 API 访问注册表（DNA、蛋白质）、库存、ELN 条目、工作流程、构建基准应用程序、查询数据仓库，实现实验室数据管理自动化。

DNAnexus cloud genomics platform. Build apps/applets, manage data (upload/download), dxpy Python SDK, run workflows, FASTQ/BAM/VCF, for genomics pipeline development and execution. DNAnexus 云基因组学平台。构建应用程序/小程序、管理数据（上传/下载）、dxpy Python SDK、运行工作流程、FASTQ/BAM/VCF，用于基因组学管道开发和执行。

Submit and manage protocols on Ginkgo Bioworks Cloud Lab (cloud.ginkgo.bio), a web-based interface for autonomous lab execution on Reconfigurable Automation Carts (RACs). Use when the user wants to run protein expression and purification (cell-free, E. coli, or Pichia), HiBiT or A280 or LabChip quantification, IVT mRNA/circRNA synthesis, thermal shift / developability assays, Echo-MS enzyme or analyte methods, SPR target onboarding, fluorescent pixel art, or otherwise interact with Ginkgo Cloud Lab services. Covers protocol selection, input preparation, pricing, and ordering workflows. 在 Ginkgo Bioworks 云实验室 (cloud.ginkgo.bio) 上提交和管理协议，这是一个基于 Web 的界面，用于在可重新配置自动化推车 (RAC) 上自主执行实验室。当用户想要运行无细胞蛋白质表达（验证或优化）、生成荧光像素艺术或与 Ginkgo Cloud Lab 服务交互时使用。涵盖协议选择、输入准备、定价和订购工作流程。

Electronic lab notebook API integration. Access notebooks, manage entries/attachments, backup notebooks, integrate with Protocols.io/Jupyter/REDCap, for programmatic ELN workflows. 电子实验室笔记本 API 集成。访问笔记本、管理条目/附件、备份笔记本、与 Protocols.io/Jupyter/REDCap 集成，实现编程式 ELN 工作流程。

Latch platform for bioinformatics workflows. Build pipelines with Latch SDK, @workflow/@task decorators, deploy serverless workflows, LatchFile/LatchDir, Nextflow/Snakemake integration. 用于生物信息学工作流程的 Latch 平台。使用 Latch SDK、@workflow/@task 装饰器构建管道，部署无服务器工作流程、LatchFile/LatchDir、Nextflow/Snakemake 集成。

Official Opentrons Protocol API for OT-2 and Flex robots. Use when writing protocols specifically for Opentrons hardware with full access to Protocol API v2 features. Best for production Opentrons protocols, official API compatibility. For multi-vendor automation or broader equipment control use pylabrobot. 适用于 OT-2 和 Flex 机器人的官方 Opentrons 协议 API。在专门为 Opentrons 硬件编写协议时使用，可以完全访问协议 API v2 功能。最适合生产 Opentrons 协议，官方 API 兼容性。对于多供应商自动化或更广泛的设备控制，请使用 pylabrobot。

Integration with protocols.io API for managing scientific protocols. This skill should be used when working with protocols.io to search, create, update, or publish protocols; manage protocol steps and materials; handle discussions and comments; organize workspaces; upload and manage files; or integrate protocols.io functionality into workflows. Applicable for protocol discovery, collaborative protocol development, experiment tracking, lab protocol management, and scientific documentation. 与 Protocols.io API 集成，用于管理科学协议。当使用protocols.io搜索、创建、更新或发布协议时，应该使用此技能；管理方案步骤和材料；处理讨论和评论；整理工作空间；上传和管理文件；或将protocols.io 功能集成到工作流程中。适用于协议发现、协作协议开发、实验跟踪、实验室协议管理和科学文档。

Vendor-agnostic lab automation framework. Use when controlling multiple equipment types (Hamilton, Tecan, Opentrons, plate readers, pumps) or needing unified programming across different vendors. Best for complex workflows, multi-vendor setups, simulation. For Opentrons-only protocols with official API, opentrons-integration may be simpler. 与供应商无关的实验室自动化框架。当控制多种设备类型（Hamilton、Tecan、Opentron、读板机、泵）或需要跨不同供应商进行统一编程时使用。最适合复杂的工作流程、多供应商设置、模拟。对于具有官方 API 的仅限 Opentrons 的协议，opentrons 集成可能更简单。

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner. 用于 4 位 LLM 压缩的激活感知权重量化，加速速度提高 3 倍，精度损失最小。当在有限的 GPU 内存上部署大型模型 (7B-70B) 时、当您需要比 GPTQ 更快的推理速度和更好的精度保留时，或者用于指令调整和多模态模型时，请使用。 MLSys 2024 最佳论文奖获得者。

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support 使用 Axolotl 微调 LLM 的专家指南 - YAML 配置、100 多个模型、LoRA/QLoRA、DPO/KTO/ORPO/GRPO、多模式支持

Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards. 使用 pass@k 指标跨 HumanEval、MBPP、MultiPL-E 和 15+ 基准评估代码生成模型。在对代码模型进行基准测试、比较编码能力、测试多语言支持或测量代码生成质量时使用。 HuggingFace 排行榜使用的 BigCode 项目的行业标准。

Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system. Anthropic 通过自我完善来训练无害人工智能的方法。两阶段方法 - 通过自我批评/修改进行监督学习，然后是 RLAIF（来自 AI 反馈的 RL）。用于安全对齐，减少有害输出，无需人工标记。为克劳德的安全系统提供动力。

GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements. GGUF 格式和 llama.cpp 量化可实现高效的 CPU/GPU 推理。在消费类硬件、Apple Silicon 上部署模型时，或者需要从 2-8 位进行灵活量化且无需 GPU 时使用。

Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning. LLM 的训练后 4 位量化，精度损失最小。当您需要 4 倍内存减少且困惑度降低 <2% 时，或者需要比 FP16 更快的推理（3-4 倍加速）时，可用于在消费类 GPU 上部署大型模型（70B、405B）。与变压器和 PEFT 集成以进行 QLoRA 微调。

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training 使用 TRL 进行 GRPO/RL 微调的专家指导，以进行推理和特定于任务的模型训练

Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers. 使用 Lightning AI 的 LitGPT 和 20 多个预训练架构（Llama、Gemma、Phi、Qwen、Mistral）实施和训练法学硕士。当需要干净的模型实现、架构的教育理解或使用 LoRA/QLoRA 进行生产微调时使用。单文件实现，无抽象层。

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU. 在 CPU、Apple Silicon 和消费类 GPU 上运行 LLM 推理，无需 NVIDIA 硬件。用于边缘部署、M1/M2/M3 Mac、AMD/Intel GPU 或 CUDA 不可用时。支持 GGUF 量化（1.5-8 位），以减少内存，与 CPU 上的 PyTorch 相比，速度提高 4-10 倍。

Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support 使用 LLaMA-Factory 微调 LLM 的专家指南 - WebUI 无代码、100 多个模型、2/3/4/5/6/8 位 QLoRA、多模式支持

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails. Meta 的 7-8B 专门用于 LLM 输入/输出过滤的调节模型。 6 个安全类别 - 暴力/仇恨、性内容、武器、物质、自残、犯罪计划。准确度为 94-95%。使用 vLLM、HuggingFace、Sagemaker 进行部署。与 NeMo 护栏集成。

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs. 通过 60 多个学术基准（MMLU、HumanEval、GSM8K、TruthfulQA、HellaSwag）评估法学硕士。在对模型质量进行基准测试、比较模型、报告学术成果或跟踪培训进度时使用。 EleutherAI、HuggingFace 和主要实验室使用的行业标准。支持 HuggingFace、vLLM、API。

State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV cache. Selective SSM with hardware-aware design. Mamba-1 (d_state=16) and Mamba-2 (d_state=128, multi-head). Models 130M-2.8B on HuggingFace. 状态空间模型的复杂度为 O(n)，而 Transformer 的复杂度为 O(n²)。推理速度提高 5 倍，百万个令牌序列，无 KV 缓存。具有硬件感知设计的选择性 SSM。 Mamba-1 (d_state=16) 和 Mamba-2 (d_state=128，多头)。 HuggingFace 上的型号 130M-2.8B。

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek. 使用 NVIDIA Megatron-Core 和高级并行策略训练大型语言模型（2B-462B 参数）。当训练模型 >1B 参数、需要最大 GPU 效率（H100 上为 47% MFU）或需要张量/管道/序列/上下文/专家并行时使用。用于 Nemotron、LLaMA、DeepSeek 的生产就绪框架。

Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime. Use when training large MoE models with FP8/INT4, needing train-inference alignment, or requiring speculative RL for maximum throughput. 为使用 Miles（一种可立即投入生产的 Slime 分支）进行企业级 RL 训练提供指导。当使用 FP8/INT4 训练大型 MoE 模型、需要训练推理对齐或需要推测 RL 以获得最大吞吐量时使用。

Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU). 教育 GPT 实现约 300 行。在 OpenWebText 上复制 GPT-2 (124M)。用于学习 Transformer 的干净、可破解的代码。安德烈·卡帕蒂着。非常适合从头开始理解 GPT 架构。在 Shakespeare (CPU) 或 OpenWebText (多 GPU) 上进行训练。

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution. Use when needing scalable evaluation on local Docker, Slurm HPC, or cloud platforms. NVIDIA's enterprise-grade platform with container-first architecture for reproducible benchmarking. 通过多后端执行，通过 18 多个工具（MMLU、HumanEval、GSM8K、安全、VLM）评估 100 多个基准的法学硕士。当需要在本地 Docker、Slurm HPC 或云平台上进行可扩展评估时使用。 NVIDIA 的企业级平台采用容器优先架构，可实现可重复的基准测试。

NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU. NVIDIA 针对 LLM 应用程序的运行时安全框架。具有越狱检测、输入/输出验证、事实检查、幻觉检测、PII 过滤、毒性检测等功能。使用 Colang 2.0 DSL 实现可编程轨。生产就绪，在 T4 GPU 上运行。

Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture. 提供使用 nnsight 和可选 NDIF 远程执行来解释和操作神经网络内部结构的指导。当需要在没有本地 GPU 资源的情况下在大规模模型 (70B+) 上运行可解释性实验或使用任何 PyTorch 架构时使用。

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing. 具有 Ray+vLLM 加速功能的高性能 RLHF 框架。用于大型模型（7B-70B+）的PPO、GRPO、RLOO、DPO训练。基于 Ray、vLLM、ZeRO-3 构建。比 DeepSpeedChat 快 2 倍，具有分布式架构和 GPU 资源共享。

Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+ TPR, <1% FPR. Fast (<2ms GPU). Multilingual (8 languages). Deploy with HuggingFace or batch processing for RAG security. Meta的86M提示注入和越狱检测器。过滤LLM应用程序的恶意提示和第三方数据。 99%+ TPR，<1% FPR。快速（<2ms GPU）。多语言（8 种语言）。使用 HuggingFace 或批处理进行部署以确保 RAG 安全。

Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior. 提供使用 pyvene 的声明性干预框架对 PyTorch 模型执行因果干预的指导。在进行因果追踪、激活修补、交换干预训练或测试有关模型行为的因果假设时使用。

RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters. RNN+Transformer 混合，具有 O(n) 推理能力。线性时间，无限上下文，无KV缓存。像 GPT 一样训练（并行），像 RNN 一样推理（顺序）。 Linux 基金会人工智能项目。在 Windows、Office、NeMo 上进行生产。 RWKV-7（2025 年 3 月）。模型多达 14B 参数。

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models. 提供使用 SAELens 训练和分析稀疏自动编码器 (SAE) 的指导，将神经网络激活分解为可解释的特征。在发现可解释的特征、分析叠加或研究语言模型中的单语义表示时使用。

Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM with prefix sharing. Powers 300,000+ GPUs at xAI, AMD, NVIDIA, and LinkedIn. 快速结构化生成并为具有 RadixAttention 前缀缓存的法学硕士提供服务。用于 JSON/正则表达式输出、受限解码、带有工具调用的代理工作流程，或者当您需要比具有前缀共享的 vLLM 快 5 倍的推理时。为 xAI、AMD、NVIDIA 和 LinkedIn 的 300,000 多个 GPU 提供支持。

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO. LLM 对齐的简单偏好优化。 DPO 的无参考替代方案，具有更好的性能（在 AlpacaEval 2.0 上+6.4 分）。无需参考模型，比DPO更高效。当需要比 DPO/PPO 更简单、更快的培训时，可用于偏好调整。

Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling. 为使用 Slime（Megatron+SGLang 框架）进行 RL 的 LLM 后期训练提供指导。在训练 GLM 模型、实施自定义数据生成工作流程或需要紧密的 Megatron-LM 集成以进行 RL 扩展时使用。

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling. 使用 NVIDIA TensorRT 优化 LLM 推理，以实现最大吞吐量和最低延迟。当您需要比 PyTorch 快 10-100 倍的推理速度时，可用于 NVIDIA GPU (A100/H100) 上的生产部署，或者用于通过量化 (FP8/INT4)、动态批处理和多 GPU 扩展来服务模型。

Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan. 使用 torchforge（将基础设施与算法分离的 Meta 库）为 PyTorch 原生代理强化学习提供指导。当您需要干净的 RL 抽象、简单的算法实验或使用 Monarch 和 TorchTitan 进行可扩展训练时使用。

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing. 使用具有 4D 并行性的 torchtitan（FSDP2、TP、PP、CP）提供 PyTorch 原生分布式 LLM 预训练。在使用 Float8、torch.compile 和分布式检查点预训练 Llama 3.1、DeepSeek V3 或自定义模型（8 到 512+ GPU）时使用。

Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments. 为使用 TransformerLens 通过 HookPoint 和激活缓存检查和操作变压器内部结构的机械可解释性研究提供指导。在逆向工程模型算法、研究注意力模式或执行激活修补实验时使用。

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers. 使用强化学习对法学硕士进行微调，TRL - SFT 用于指令调整，DPO 用于偏好调整，PPO/GRPO 用于奖励优化，以及奖励模型训练。在需要 RLHF 时使用，根据偏好调整模型，或根据人类反馈进行训练。与 HuggingFace 变形金刚一起使用。

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization 使用 Unsloth 进行快速微调的专家指导 - 训练速度提高 2-5 倍，内存减少 50-80%，LoRA/QLoRA 优化

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends. 为使用 verl (Volcano Engine RL) 进行强化学习培训法学硕士提供指导。在通过灵活的基础设施后端大规模实施 RLHF、GRPO、PPO 或其他 RL 算法以进行 LLM 后期培训时使用。

Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism. 使用 vLLM 的 PagedAttention 和连续批处理为法学硕士提供高吞吐量服务。在部署生产 LLM API、优化推理延迟/吞吐量或为 GPU 内存有限的模型提供服务时使用。支持 OpenAI 兼容端点、量化 (GPTQ/AWQ/FP8) 和张量并行性。

PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation. 用于音频生成的 PyTorch 库，包括文本到音乐 (MusicGen) 和文本到声音 (AudioGen)。当您需要从文本描述生成音乐、创建声音效果或执行旋律条件音乐生成时使用。

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance. 连接冻结图像编码器和法学硕士的视觉语言预训练框架。当您需要图像字幕、视觉问答、图像文本检索或多模式聊天以及最先进的零样本性能时使用。

OpenAI's model connecting vision and language. Enables zero-shot image classification, image-text matching, and cross-modal retrieval. Trained on 400M image-text pairs. Use for image search, content moderation, or vision-language tasks without fine-tuning. Best for general-purpose image understanding. OpenAI 连接视觉和语言的模型。实现零样本图像分类、图像文本匹配和跨模式检索。使用 4 亿个图像-文本对进行训练。用于图像搜索、内容审核或视觉语言任务，无需微调。最适合通用图像理解。

Evaluates NVIDIA Cosmos Policy on LIBERO and RoboCasa simulation environments. Use when setting up cosmos-policy for robot manipulation evaluation, running headless GPU evaluations with EGL rendering, or profiling inference latency on cluster or local GPU machines. 评估 LIBERO 和 RoboCasa 模拟环境中的 NVIDIA Cosmos 政策。在为机器人操作评估设置 cosmos-policy、使用 EGL 渲染运行无头 GPU 评估或分析集群或本地 GPU 计算机上的推理延迟时使用。

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis. 大型语言和视觉助手。实现视觉指令调整和基于图像的对话。将 CLIP 视觉编码器与 Vicuna/LLaMA 语言模型相结合。支持多轮图像聊天、可视化问答、跟随指令。用于视觉语言聊天机器人或图像理解任务。最适合对话式图像分析。

Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging norm stats and GPU memory issues. 使用 JAX 或 PyTorch 后端微调和服务物理智能 OpenPI 模型（pi0、pi0-fast、pi0.5），以在 ALOHA、DROID 和 LIBERO 环境中进行机器人策略推理。在将 pi0 模型调整为自定义数据集、将 JAX 检查点转换为 PyTorch、运行策略推理服务器或调试规范统计数据和 GPU 内存问题时使用。

Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues. 通过连续动作头、LoRA 适应以及 LIBERO 模拟和 ALOHA 现实世界设置的 FiLM 调节，微调和评估机器人动作生成的 OpenVLA-OFT 和 OpenVLA-OFT+ 策略。在重现 OpenVLA-OFT 论文结果、训练自定义 VLA 操作头（L1 或扩散）、部署 ALOHA 的服务器-客户端推理或调试规范化、LoRA 合并和跨 GPU 问题时使用。

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image. 零样本传输图像分割的基础模型。当您需要使用点、框或蒙版作为提示来分割图像中的任何对象时使用，或者自动生成图像中的所有对象蒙版。

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines. 通过 HuggingFace 扩散器使用稳定扩散模型生成最先进的文本到图像。在根据文本提示生成图像、执行图像到图像转换、修复或构建自定义扩散管道时使用。

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR. OpenAI 的通用语音识别模型。支持99种语言、转录、翻译成英文、语言识别。六种模型大小，从小（39M 参数）到大（1550M 参数）。用于语音转文本、播客转录或多语言音频处理。最适合强大的多语言 ASR。

Provides guidance for automatically evolving and optimizing AI agents across any domain using LLM-driven evolution algorithms. Use when building self-improving agents, optimizing agent prompts and skills against benchmarks, or implementing automated agent evaluation loops. 提供使用 LLM 驱动的进化算法在任何领域自动进化和优化 AI 代理的指导。在构建自我改进代理、根据基准优化代理提示和技能或实施自动代理评估循环时使用。

This skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring specialized algorithms beyond standard ML approaches. Particularly suited for univariate and multivariate time series analysis with scikit-learn compatible APIs. 该技能应用于时间序列机器学习任务，包括分类、回归、聚类、预测、异常检测、分割和相似性搜索。当处理时态数据、顺序模式或时间索引观察需要标准 ML 方法之外的专用算法时使用。特别适合使用 scikit-learn 兼容 API 进行单变量和多变量时间序列分析。

Autonomous AI agent platform for building and deploying continuous agents. Use when creating visual workflow agents, deploying persistent autonomous agents, or building complex multi-step AI automation systems. 用于构建和部署连续代理的自主人工智能代理平台。在创建可视化工作流代理、部署持久自主代理或构建复杂的多步骤 AI 自动化系统时使用。

Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers. 将 LLM 量化为 8 位或 4 位，可减少 50-75% 的内存，同时将精度损失降至最低。当 GPU 内存有限、需要适应更大的模型或想要更快的推理时使用。支持 INT8、NF4、FP4 格式、QLoRA 训练和 8 位优化器。与 HuggingFace 变形金刚一起使用。

Guides researchers through structured ideation frameworks to discover high-impact research directions. Use when exploring new problem spaces, pivoting between projects, or seeking novel angles on existing work. 指导研究人员通过结构化的构思框架来发现高影响力的研究方向。在探索新的问题空间、在项目之间进行调整或在现有工作中寻找新的角度时使用。

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source projects. 用于人工智能应用的开源嵌入数据库。存储嵌入和元数据，执行矢量和全文搜索，按元数据过滤。简单的 4 功能 API。从笔记本扩展到生产集群。用于语义搜索、RAG 应用程序或文档检索。最适合本地开发和开源项目。

Compiles any research input — PDF papers, GitHub repositories, experiment logs, code directories, or raw notes — into a complete Agent-Native Research Artifact (ARA) with cognitive layer (claims, concepts, heuristics), physical layer (configs, code stubs), exploration graph, and grounded evidence. Use when ingesting a paper or codebase into a structured, machine-executable knowledge package, building an ARA from scratch, or converting research outputs into a falsifiable, agent-traversable form. 将任何研究输入（PDF 论文、GitHub 存储库、实验日志、代码目录或原始笔记）编译为完整的 Agent-Native 研究工件 (ARA)，其中包含认知层（声明、概念、启发式）、物理层（配置、代码存根）、探索图和基础证据。在将论文或代码库提取到结构化的机器可执行知识包、从头开始构建 ARA 或将研究输出转换为可证伪的、代理可遍历的形式时使用。

Run a multi-perspective Mind Council deliberation on any question, decision, or creative challenge. Use this skill whenever the user wants diverse viewpoints, needs help making a tough decision, asks for a council/panel/board discussion, wants to explore a problem from multiple angles, requests devil's advocate analysis, or says things like "what would different experts think about this", "help me think through this from all sides", "council mode", "mind council", or "deliberate on this". Also trigger when the user faces a dilemma, trade-off, or complex choice with no obvious answer. 对任何问题、决定或创造性挑战进行多角度的思想委员会审议。每当用户想要不同的观点、需要帮助做出艰难的决定、要求进行理事会/小组/理事会讨论、想要从多个角度探讨问题、要求魔鬼代言人分析、或者说“不同的专家会怎么看待这个问题”、“帮助我从各个方面思考这个问题”、“理事会模式”、“心灵理事会”或“仔细考虑这个问题”之类的事情时，就可以使用此技能。当用户面临困境、权衡或没有明显答案的复杂选择时也会触发。

Applies cognitive science frameworks for creative thinking to CS and AI research ideation. Use when seeking genuinely novel research directions by leveraging combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded creative strategies. 将创造性思维的认知科学框架应用于计算机科学和人工智能研究构思。通过利用组合创造力、类比推理、约束操纵和其他基于经验的创造性策略来寻求真正新颖的研究方向时使用。

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution. 用于自主人工智能协作的多代理编排框架。当构建专门代理团队共同处理复杂任务时，当您需要基于角色的代理与内存协作时，或者需要顺序/分层执行的生产工作流程时，请使用。构建时不依赖 LangChain，以实现精简、快速的执行。