[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skills:sce6":3,"skills:pro6":50,"skills:creators9":95,"skills:stats":193,"skills:all:stars:10818:":195},[4,15,22,29,36,43],{"categoryNo":5,"taxonomyType":6,"taxonomyTypeDesc":7,"categoryName":8,"categoryNameEn":9,"description":10,"level":11,"skillCount":12,"parentNo":13,"parentName":13,"children":14},"01JBSEEDCAT000000000000022","SCENARIO","使用场景","智能体与自动化","Agents & Automation","Skill、MCP、智能体与工作流自动化",1,39458,null,[],{"categoryNo":16,"taxonomyType":6,"taxonomyTypeDesc":7,"categoryName":17,"categoryNameEn":18,"description":19,"level":11,"skillCount":20,"parentNo":13,"parentName":13,"children":21},"01JBSEEDCAT000000000000006","软件开发","Software Development","编码、测试、部署等软件研发任务",26531,[],{"categoryNo":23,"taxonomyType":6,"taxonomyTypeDesc":7,"categoryName":24,"categoryNameEn":25,"description":26,"level":11,"skillCount":27,"parentNo":13,"parentName":13,"children":28},"01JBSEEDCAT000000000000014","设计创意","Design & Creative","视觉设计、品牌与多媒体创作",13569,[],{"categoryNo":30,"taxonomyType":6,"taxonomyTypeDesc":7,"categoryName":31,"categoryNameEn":32,"description":33,"level":11,"skillCount":34,"parentNo":13,"parentName":13,"children":35},"01JBSEEDCAT000000000000001","办公效率","Office Productivity","日常办公文档、表格、演示与沟通类任务",6289,[],{"categoryNo":37,"taxonomyType":6,"taxonomyTypeDesc":7,"categoryName":38,"categoryNameEn":39,"description":40,"level":11,"skillCount":41,"parentNo":13,"parentName":13,"children":42},"01JBSEEDCAT000000000000017","内容创作","Content Creation","写作、文案与营销内容生产",6029,[],{"categoryNo":44,"taxonomyType":6,"taxonomyTypeDesc":7,"categoryName":45,"categoryNameEn":46,"description":47,"level":11,"skillCount":48,"parentNo":13,"parentName":13,"children":49},"01JBSEEDCAT000000000000020","研究与学习","Research & Learning","资料检索、知识整理与学习辅导",5531,[],[51,60,67,74,81,88],{"categoryNo":52,"taxonomyType":53,"taxonomyTypeDesc":54,"categoryName":55,"categoryNameEn":56,"description":57,"level":11,"skillCount":58,"parentNo":13,"parentName":13,"children":59},"01JBSEEDCAT000000000000030","PROFESSION","适用职业","开发工程师","Software Engineer","从事软件编码、测试与运维的工程师",13274,[],{"categoryNo":61,"taxonomyType":53,"taxonomyTypeDesc":54,"categoryName":62,"categoryNameEn":63,"description":64,"level":11,"skillCount":65,"parentNo":13,"parentName":13,"children":66},"01JBSEEDCAT000000000000032","设计师","Designer","视觉、交互与品牌设计人员",7775,[],{"categoryNo":68,"taxonomyType":53,"taxonomyTypeDesc":54,"categoryName":69,"categoryNameEn":70,"description":71,"level":11,"skillCount":72,"parentNo":13,"parentName":13,"children":73},"01JBSEEDCAT000000000000033","数据分析师","Data Analyst","数据分析、报表与商业智能人员",4988,[],{"categoryNo":75,"taxonomyType":53,"taxonomyTypeDesc":54,"categoryName":76,"categoryNameEn":77,"description":78,"level":11,"skillCount":79,"parentNo":13,"parentName":13,"children":80},"01JBSEEDCAT000000000000036","内容创作者","Content Creator","作者、编辑与自媒体创作者",3940,[],{"categoryNo":82,"taxonomyType":53,"taxonomyTypeDesc":54,"categoryName":83,"categoryNameEn":84,"description":85,"level":11,"skillCount":86,"parentNo":13,"parentName":13,"children":87},"01JBSEEDCAT000000000000035","市场营销","Marketer","市场推广、增长与品牌运营人员",3446,[],{"categoryNo":89,"taxonomyType":53,"taxonomyTypeDesc":54,"categoryName":90,"categoryNameEn":91,"description":92,"level":11,"skillCount":93,"parentNo":13,"parentName":13,"children":94},"01JBSEEDCAT000000000000037","律师与法务","Legal","合同、合规与法律事务人员",3316,[],{"records":96,"total":191,"current":11,"size":192},[97,111,120,130,140,150,160,169,179],{"accountNo":98,"login":99,"displayName":100,"avatarUrl":101,"accountType":102,"accountTypeDesc":103,"profileDescription":104,"skillCount":105,"repoCount":106,"professionCount":107,"scenarioCount":108,"totalStars":109,"updatedAt":110},"01M698VYDXZ6JW5WH3CRV650R8","microsoft","Microsoft","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6154722?v=4","ORGANIZATION","组织","Open source projects and samples from Microsoft",4160,229,11,7,828772,"2026-11-13 04:10:52",{"accountNo":112,"login":113,"displayName":113,"avatarUrl":114,"accountType":102,"accountTypeDesc":103,"profileDescription":115,"skillCount":116,"repoCount":117,"professionCount":108,"scenarioCount":108,"totalStars":118,"updatedAt":119},"01M698P3FSAZ6B4R928ZF19KPT","openclaw","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F252820863?v=4","Your personal, open source AI assistant.",362,32,437711,"2026-11-13 03:02:05",{"accountNo":121,"login":122,"displayName":123,"avatarUrl":124,"accountType":102,"accountTypeDesc":103,"profileDescription":13,"skillCount":125,"repoCount":126,"professionCount":108,"scenarioCount":127,"totalStars":128,"updatedAt":129},"01M6AC29YGCR4S2NTJFV0G8B81","openai","OpenAI","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F14957082?v=4",171,23,6,342695,"2026-11-13 04:03:41",{"accountNo":131,"login":132,"displayName":133,"avatarUrl":134,"accountType":102,"accountTypeDesc":103,"profileDescription":13,"skillCount":135,"repoCount":136,"professionCount":137,"scenarioCount":108,"totalStars":138,"updatedAt":139},"01M6AC83E6K4FDKK7HZ8WNMQKE","apache","The Apache Software Foundation","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F47359?v=4",420,75,8,298176,"2026-11-13 03:14:28",{"accountNo":141,"login":142,"displayName":143,"avatarUrl":144,"accountType":102,"accountTypeDesc":103,"profileDescription":145,"skillCount":146,"repoCount":147,"professionCount":107,"scenarioCount":108,"totalStars":148,"updatedAt":149},"01M698VYEFDM0VJBZDYYJ2T83S","google","Google","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1342004?v=4","Google ❤️ Open Source",535,48,242366,"2026-11-13 04:17:09",{"accountNo":151,"login":152,"displayName":153,"avatarUrl":154,"accountType":102,"accountTypeDesc":103,"profileDescription":155,"skillCount":156,"repoCount":157,"professionCount":107,"scenarioCount":108,"totalStars":158,"updatedAt":159},"01M6AC83DS133TPXFSP1DHNCC6","github","GitHub","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F9919?v=4","How people build software.",940,28,238441,"2026-11-13 03:57:35",{"accountNo":161,"login":162,"displayName":163,"avatarUrl":164,"accountType":102,"accountTypeDesc":103,"profileDescription":13,"skillCount":165,"repoCount":166,"professionCount":107,"scenarioCount":108,"totalStars":167,"updatedAt":168},"01M697XGDPXJAVQ8391V2BCYTE","anthropics","Anthropic","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F76263028?v=4",674,17,231826,"2026-11-13 04:18:26",{"accountNo":170,"login":171,"displayName":172,"avatarUrl":173,"accountType":102,"accountTypeDesc":103,"profileDescription":13,"skillCount":174,"repoCount":175,"professionCount":176,"scenarioCount":108,"totalStars":177,"updatedAt":178},"01M6AC83EPDGBFN19MWYRREJD0","Significant-Gravitas","AutoGPT","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F130738209?v=4",86,2,10,188014,"2026-11-13 02:38:59",{"accountNo":180,"login":181,"displayName":182,"avatarUrl":183,"accountType":184,"accountTypeDesc":185,"profileDescription":186,"skillCount":187,"repoCount":188,"professionCount":127,"scenarioCount":108,"totalStars":189,"updatedAt":190},"01M6DJCCDDG53MTYYRTXAXFG17","garrytan","Garry Tan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F19957?v=4","USER","个人作者","Writes software, dreams. ",324,3,174614,"2026-11-13 04:05:07",1926,9,{"skillCount":194,"professionCount":107,"scenarioCount":108,"creatorCount":191},196852,{"records":196,"total":194,"current":597,"size":598},[197,233,255,275,298,319,344,371,392,416,441,465,485,505,526,540,559,579],{"skillNo":198,"name":199,"nameZh":13,"description":200,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":209,"categories":210,"keywords":211,"publishedAt":232},"01M6XGKR3FTB5WA8AJRXYACKD9","fluid-dynamics","Computational fluid dynamics — Navier-Stokes solvers, lid-driven cavity, channel flow, vortex methods, turbulence statistics, drag\u002Flift computation. Spectral and finite-difference methods for incompressible and compressible flows.","fadrienne\u002FMyGitHub","fadrienne","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F197874571?v=4","01M6X64V8VWCRSMFZ4EPS2BWXG","Francesca Adrienne","01M6X64VD5MW4F9PHDWQKS7ZAF",0,"MIT","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fphysics\u002Ffluid-dynamics\u002FSKILL.md",[],[212,213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231],"fluid","fluid dynamics","dynamics","computational fluid dynamics","methods","numpy","scipy","physics","computational","statistics","synthetic","lift","drag","CFD","turbulence","spectral","flows","for","computation","Incompressible Flow","2026-11-10 10:12:30",{"skillNo":234,"name":235,"nameZh":13,"description":236,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":237,"categories":238,"keywords":239,"publishedAt":232},"01M6XGKJH38S0H33N3F3ER8V4D","dynamical-systems","Analyze nonlinear dynamical systems — phase portraits, fixed points, stability analysis, bifurcation diagrams, Poincare sections, Lyapunov exponents, and chaos detection. Use for any autonomous or non-autonomous ODE system where qualitative behavior matters.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fphysics\u002Fdynamical-systems\u002FSKILL.md",[],[240,241,242,243,244,245,246,247,248,217,218,219,249,250,222,229,251,252,253,254],"systems","Dynamical Systems","analysis","matplotlib","system","analyze","autonomous","use","diagrams","portrait","ode","phase","Phase Portrait","Bifurcation","Nonlinear",{"skillNo":256,"name":257,"nameZh":13,"description":258,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":259,"categories":260,"keywords":261,"publishedAt":232},"01M6XGKDDH204S2HW7H2QEKJ4Q","dimensional-analysis","Automated dimensional analysis — Buckingham Pi theorem, non-dimensionalization, unit validation with pint, and characteristic scale estimation. Use before any physics computation to verify consistency and reduce parameter space.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fphysics\u002Fdimensional-analysis\u002FSKILL.md",[],[242,262,263,264,265,247,266,217,218,267,219,222,268,269,270,271,272,230,273,274],"dimensional analysis","validation","verify","pi","units","estimation","unit","with","consistency","reduce","space","Buckingham Pi","Non-dimensionalization",{"skillNo":276,"name":277,"nameZh":13,"description":278,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":279,"categories":280,"keywords":282,"publishedAt":232},"01M6XGK868PA40EZ8EZRGMM7P0","conservation-law-discovery","Discover conserved quantities and symmetries from trajectory data. Identifies energy, momentum, angular momentum, and custom invariants using neural networks and symbolic methods. Inspired by Noether's theorem.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fphysics\u002Fconservation-law-discovery\u002FSKILL.md",[281],{"categoryNo":89,"categoryName":90,"taxonomyType":53,"taxonomyTypeDesc":54},[283,284,285,286,287,288,289,216,217,218,219,290,222,291,292,293,294,295,296,297],"discovery","conservation","conservation law discovery","neural networks","angular","networks","Neural","discover","trajectory","custom","energy","Conservation Laws","Noether","Symmetry","Physics Discovery",{"skillNo":299,"name":300,"nameZh":13,"description":301,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":302,"categories":303,"keywords":304,"publishedAt":232},"01M6XGK49W4DJ2KA8CDCDGRH56","bayesian-inference","Bayesian parameter estimation with MCMC (emcee) and probabilistic programming (PyMC). Posterior distributions, corner plots, model evidence, convergence diagnostics. Use when you need full posterior distributions, not just point estimates.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fphysics\u002Fbayesian-inference\u002FSKILL.md",[],[305,306,307,308,309,247,310,217,218,267,219,222,311,312,313,314,315,316,317,318],"inference","Bayesian","bayesian inference","model","just","evidence","you","uncertainty","point","when","MCMC","Probabilistic Programming","Posterior","emcee",{"skillNo":320,"name":321,"nameZh":13,"description":322,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":323,"categories":324,"keywords":325,"publishedAt":232},"01M6XGJZG4BHDS0V1HVRMHJ74N","autoregressive-neural-pde-solver","Training patterns for autoregressive neural PDE solvers (FNO, DeepONet, CNO). Covers rollout training, noise injection for stability, multi-component loss functions (H1, frequency-sensitive, boundary-aware), per-channel normalization for coupled multi-variable systems, and the PDEBench nRMSE metric. Use when training any neural operator that predicts time-dependent PDE solutions.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fphysics\u002Fautoregressive-neural-pde-solver\u002FSKILL.md",[],[289,326,327,328,329,330,331,332,333,334,335,336,219,222,337,314,338,339,340,341,342,343],"pde","solver","autoregressive neural pde solver","pde solver","component","training","operator","injection","rollout","patterns","time","metric","solutions","Neural Operator","Autoregressive","FNO","Loss Function","DeepONet",{"skillNo":345,"name":346,"nameZh":13,"description":347,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":348,"categories":349,"keywords":353,"publishedAt":232},"01M6XGJGW42MAZHZ1A2TJF0ACQ","skill-installer","Install or remove third-party openscience skills from a public git repository. Use when the user says \"add this skill \u003Curl>\", \"install skill \u003Curl>\", or \"remove skill \u003Cnamespace>\". The skill runs locally via `openscience skill add|list|remove`, fetches the repo, runs a 6-layer safety gate (regex + server-side Haiku classifier), prompts the user to confirm, then writes the skills to ~\u002F.openscience\u002Finstalled-skills\u002F and uploads to the dashboard for cross-machine sync.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fother\u002Fskill-installer\u002FSKILL.md",[350],{"categoryNo":351,"categoryName":352,"taxonomyType":6,"taxonomyTypeDesc":7},"01JBSEEDCAT000000000000023","Skill与智能体开发",[354,355,356,357,358,359,360,361,362,247,363,364,365,366,367,368,369,370,229,314],"skill installer","installer","dashboard","sync","server","repository","haiku","repo","remove","classifier","uploads","server-side","install skill","regex","via","machine","URL",{"skillNo":372,"name":373,"nameZh":13,"description":374,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":375,"categories":376,"keywords":377,"publishedAt":232},"01M6XGJ1R0BJPK6D9DTTVVX8CV","iso-13485-certification","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.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fother\u002Fiso-13485-certification\u002FSKILL.md",[],[378,379,380,381,382,383,242,244,384,247,385,386,387,388,222,389,390,229,391,314],"certification","iso 13485 certification","iso","ISO 13485","documentation","toolkit","work","files","requirements","medical","gap analysis","missing","eu mdr","Existing",{"skillNo":393,"name":394,"nameZh":13,"description":395,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":396,"licenseFileUrl":13,"sourceHtmlUrl":397,"categories":398,"keywords":402,"publishedAt":232},"01M6XGHXH1PMV2B2WET5SZCM5Y","hugging-face-trackio","Track and visualize ML training experiments with Trackio. Use when logging metrics during training (Python API) or retrieving\u002Fanalyzing logged metrics (CLI). Supports real-time dashboard visualization, HF Space syncing, and JSON output for automation.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fother\u002Fhugging-face-trackio\u002FSKILL.md",[399],{"categoryNo":400,"categoryName":401,"taxonomyType":6,"taxonomyTypeDesc":7},"01JBSEEDCAT000000000000013","分析与可视化",[403,404,405,406,407,408,409,331,410,411,412,413,247,336,222,414,415,269,229,314],"hugging face","hugging face trackio","face","python","visualize","automation","ml","visualization","HuggingFace","real-time","api","Experiment Tracking","output",{"skillNo":417,"name":418,"nameZh":13,"description":419,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":420,"categories":421,"keywords":425,"publishedAt":232},"01M6XGHPADN70PC87TD9HFPAE4","hugging-face-jobs","This skill should be used when users want to run any workload on Hugging Face Jobs infrastructure. Covers UV scripts, Docker-based jobs, hardware selection, cost estimation, authentication with tokens, secrets management, timeout configuration, and result persistence. Designed for general-purpose compute workloads including data processing, inference, experiments, batch jobs, and any Python-based tasks. Should be invoked for tasks involving cloud compute, GPU workloads, or when users mention running jobs on Hugging Face infrastructure without local setup.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fother\u002Fhugging-face-jobs\u002FSKILL.md",[422],{"categoryNo":423,"categoryName":424,"taxonomyType":6,"taxonomyTypeDesc":7},"01JBSEEDCAT000000000000011","运维与部署",[403,426,427,405,406,428,429,430,331,431,432,433,434,435,436,437,438,439,440,229],"jobs","hugging face jobs","experiments","cost estimation","infrastructure","run","authentication","batch","vLLM","uv","general","selection","compute","local","tokens",{"skillNo":442,"name":443,"nameZh":13,"description":444,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":445,"categories":446,"keywords":447,"publishedAt":232},"01M6XGHGNHHYCKV5CH4EQMBBGQ","get-available-resources","This skill should be used at the start of any computationally intensive scientific task to detect and report available system resources (CPU cores, GPUs, memory, disk space). It creates a JSON file with resource information and strategic recommendations that inform computational approach decisions such as whether to use parallel processing (joblib, multiprocessing), out-of-core computing (Dask, Zarr), GPU acceleration (PyTorch, JAX), or memory-efficient strategies. Use this skill before running analyses, training models, processing large datasets, or any task where resource constraints matter.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fother\u002Fget-available-resources\u002FSKILL.md",[],[448,449,450,451,452,331,453,454,455,456,220,457,222,458,459,460,461,462,463,464],"get available resources","get","resources","memory","pytorch","start","disk","parallel","strategies","recommendations","scientific","processing","CPU","GPU Acceleration","information","at","running",{"skillNo":466,"name":467,"nameZh":13,"description":468,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":469,"categories":470,"keywords":472,"publishedAt":232},"01M6XGHB93PBRQP236GW9HW63G","weights-and-biases","Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Fweights-and-biases\u002FSKILL.md",[471],{"categoryNo":400,"categoryName":401,"taxonomyType":6,"taxonomyTypeDesc":7},[473,428,452,331,410,474,475,476,308,477,478,336,222,414,479,480,481,482,483,484],"weights and biases","registry","logging","tuning","experiment","optimize","Model Registry","Hyperparameter Tuning","WandB","Real-Time Visualization","track","automatic",{"skillNo":486,"name":487,"nameZh":13,"description":488,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":489,"categories":490,"keywords":491,"publishedAt":232},"01M6XGH68H3D8EMZPBE2GW6A3Y","verl-rl-training","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.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Fverl\u002FSKILL.md",[],[331,492,493,494,495,430,496,497,498,499,500,501,247,502,222,503,463,504,229,314],"rl","verl","verl rl training","llms","Distributed Training","Post-Training","RLHF","PPO","Transformers","learning","algorithms","volcano","implementing",{"skillNo":506,"name":507,"nameZh":13,"description":508,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":509,"categories":510,"keywords":511,"publishedAt":232},"01M6XGGYBC7CYVVTCEN8NSG1AK","unsloth-fine-tuning","Fast LLM fine-tuning with Unsloth - 2-5x faster training, 50-80% less VRAM. Use for single-GPU LoRA\u002FQLoRA SFT, GRPO\u002FRL reasoning training, vision\u002FTTS fine-tuning, and GGUF export to Ollama\u002FvLLM\u002Fllama.cpp. Supports 300+ models including Llama, Qwen, Gemma, DeepSeek, Mistral, Phi, and gpt-oss.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Funsloth\u002FSKILL.md",[],[512,476,513,514,451,331,515,500,516,517,518,519,520,222,521,522,523,524,229,525],"Unsloth","fine tuning","unsloth fine tuning","cpp","phi","tts","ollama","export","vision","GGUF","llama.cpp","TRL","vram","single",{"skillNo":527,"name":528,"nameZh":13,"description":529,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":530,"categories":531,"keywords":532,"publishedAt":232},"01M6XGGS9W1PS6K7XNXNJJGCB3","fine-tuning-with-trl","Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO\u002FGRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Ftrl-fine-tuning\u002FSKILL.md",[],[476,513,533,523,269,495,331,534,497,500,501,308,535,536,537,538,222,539,229,314],"fine tuning with trl","Optimization","feedback","alignment","align","human","Accelerate",{"skillNo":541,"name":542,"nameZh":13,"description":543,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":544,"categories":545,"keywords":546,"publishedAt":232},"01M6XGGMERHPCT9DDV6D35ZT0E","transformer-lens-interpretability","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.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Ftransformer-lens\u002FSKILL.md",[],[547,548,549,550,551,428,331,552,242,553,308,554,335,502,555,222,556,557,558,314],"lens","Transformer","Interpretability","transformer lens","transformer lens interpretability","reverse-engineering","patching","activation","caching","circuit analysis","Activation Patching","inspect",{"skillNo":560,"name":561,"nameZh":13,"description":562,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":563,"categories":564,"keywords":566,"publishedAt":232},"01M6XGGEPYZPATQ5CQCKHHNB8X","training-data-pipeline","Build training datasets for LLM specialization from production data, frontier model distillation, and synthetic bootstrapping. Use when formatting production logs into SFT data, distilling from frontier APIs, or preparing data for fine-tuning. Covers JSONL formatting, data quality validation, deduplication, and train\u002Feval splitting.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Ftraining-data-pipeline\u002FSKILL.md",[565],{"categoryNo":68,"categoryName":69,"taxonomyType":53,"taxonomyTypeDesc":54},[331,567,568,569,570,571,572,122,573,500,574,476,308,575,247,576,222,577,578,229,314],"data pipeline","data","pipeline","Training Data","training data pipeline","synthetic data","logs","deduplication","eval","formatting","distillation","jsonl",{"skillNo":580,"name":581,"nameZh":13,"description":582,"descriptionZh":13,"repositoryFullName":201,"ownerLogin":202,"ownerAvatarUrl":203,"ownerAccountNo":204,"ownerDisplayName":205,"repositoryNo":206,"starsCount":207,"licenseSpdxId":208,"licenseFileUrl":13,"sourceHtmlUrl":583,"categories":584,"keywords":585,"publishedAt":232},"01M6XGG8ATSDHVR8TFDBQCEW10","distributed-llm-pretraining-torchtitan","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.","https:\u002F\u002Fgithub.com\u002Ffadrienne\u002FMyGitHub\u002Fblob\u002Fmain\u002Fopenscience\u002Fbackend\u002Fcli\u002Fskills\u002Fml-training\u002Ftorchtitan\u002FSKILL.md",[],[586,587,588,589,590,452,331,591,496,569,308,455,592,222,593,594,595,596,463,314],"distributed","distributed llm pretraining torchtitan","torchtitan","Pretraining","architecture","compile","context","Model Architecture","FSDP2","native","pp",10818,18]