[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-optimization-from-data-orchestrator":3,"mdc-s1jy5q-key":43,"related-repo-nvidia-optimization-from-data-orchestrator":306,"related-org-nvidia-optimization-from-data-orchestrator":392},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":38,"sourceUrl":41,"mdContent":42},"optimization-from-data-orchestrator","orchestrate optimization data and solving","Coordinate uploaded data plus a natural-language question into interpretation, clarification, cuOpt solve, and a user-facing answer.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,22],{"name":13,"slug":14,"type":15},"Data Engineering","data-engineering","tag",{"name":17,"slug":18,"type":15},"Automation","automation",{"name":20,"slug":21,"type":15},"Optimization","optimization",{"name":9,"slug":8,"type":15},462,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fcuopt-examples","2026-07-14T05:32:59.680482","Apache-2.0",81,[29,30,31,32,33,21,34,35,36,37],"combinatorial-optimization","gpu","linear-programming","mixed-integer-programming","operations-research","optimization-algorithms","route-optimization","traveling-salesman-problem","vehicle-routing-problem",{"repoUrl":24,"stars":23,"forks":27,"topics":39,"description":40},[29,30,31,32,33,21,34,35,36,37],"NVIDIA cuOpt examples for decision optimization","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fcuopt-examples\u002Ftree\u002FHEAD\u002Fcuopt_on_nemoclaw\u002Fopenclaw-skills\u002Foptimization-from-data-orchestrator","---\nname: optimization-from-data-orchestrator\nversion: \"26.06.01\"\ndescription: Coordinate uploaded data plus a natural-language question into interpretation, clarification, cuOpt solve, and a user-facing answer.\nlicense: Apache-2.0\nmetadata:\n  author: NVIDIA cuOpt Team\n  tags:\n    - cuopt\n    - nemoclaw\n    - orchestration\norigin: skill-evolution\n---\n\n# Optimization From Data Orchestrator\n\nTop-level coordinator when a user provides tabular data and wants a\nconstructive plan (schedule, assign, allocate, route — any wording).\n\n**NemoClaw:** read `cuopt-sandbox\u002Freferences\u002Factivation.md` for skill\norder and cuOpt-before-heuristic rules.\n\n## When to use\n\n**Both** must hold:\n\n- tabular data provided or expected (CSV, etc.)\n- user wants a **plan from that data** (any phrasing; minimize\u002Foptimal not required)\n\nSkip for **analytics-only** requests (summarize, chart, filter), fully\npre-specified math outside this flow, or explicit replayable\u002Fauditable path.\n\n## Sequence\n\n**Step 0 (NemoClaw — do not skip):** See `cuopt-sandbox` — probe → env →\nsmoke. No schedule\u002Fheuristic output before smoke passes.\n\n1. **`optimization-intent-router`** — optimization family (LP\u002FMILP\u002FQP\u002Frouting)\n2. **`optimization-mode-router`** — only if replay\u002Faudit\u002Fexport signals\n3. **`tabular-optimization-ingestion`** — table roles (interpretation only)\n4. **`cuopt-model-mapper`** — clarify if needed, map to cuOpt, solve\n\nHandoffs after step 4:\n\n- LP \u002F MILP \u002F QP → `numerical-optimization-formulation` → `cuopt-numerical-optimization-api-python`\n- Routing → `routing-formulation` → `cuopt-routing-api-python`\n\n## Guardrails\n\n- First solver that emits assignments\u002Fschedules must be **cuOpt** after step 0\n- Ingestion steps do not authorize heuristic or greedy stand-ins\n- Do not skip intent classification; do not use cuOpt for pure analytics\n- One focused clarification beats a long questionnaire\n",{"data":44,"body":53},{"name":4,"version":45,"description":6,"license":26,"metadata":46,"origin":52},"26.06.01",{"author":47,"tags":48},"NVIDIA cuOpt Team",[49,50,51],"cuopt","nemoclaw","orchestration","skill-evolution",{"type":54,"children":55},"root",[56,64,70,90,97,107,129,141,147,165,225,230,270,276],{"type":57,"tag":58,"props":59,"children":60},"element","h1",{"id":4},[61],{"type":62,"value":63},"text","Optimization From Data Orchestrator",{"type":57,"tag":65,"props":66,"children":67},"p",{},[68],{"type":62,"value":69},"Top-level coordinator when a user provides tabular data and wants a\nconstructive plan (schedule, assign, allocate, route — any wording).",{"type":57,"tag":65,"props":71,"children":72},{},[73,79,81,88],{"type":57,"tag":74,"props":75,"children":76},"strong",{},[77],{"type":62,"value":78},"NemoClaw:",{"type":62,"value":80}," read ",{"type":57,"tag":82,"props":83,"children":85},"code",{"className":84},[],[86],{"type":62,"value":87},"cuopt-sandbox\u002Freferences\u002Factivation.md",{"type":62,"value":89}," for skill\norder and cuOpt-before-heuristic rules.",{"type":57,"tag":91,"props":92,"children":94},"h2",{"id":93},"when-to-use",[95],{"type":62,"value":96},"When to use",{"type":57,"tag":65,"props":98,"children":99},{},[100,105],{"type":57,"tag":74,"props":101,"children":102},{},[103],{"type":62,"value":104},"Both",{"type":62,"value":106}," must hold:",{"type":57,"tag":108,"props":109,"children":110},"ul",{},[111,117],{"type":57,"tag":112,"props":113,"children":114},"li",{},[115],{"type":62,"value":116},"tabular data provided or expected (CSV, etc.)",{"type":57,"tag":112,"props":118,"children":119},{},[120,122,127],{"type":62,"value":121},"user wants a ",{"type":57,"tag":74,"props":123,"children":124},{},[125],{"type":62,"value":126},"plan from that data",{"type":62,"value":128}," (any phrasing; minimize\u002Foptimal not required)",{"type":57,"tag":65,"props":130,"children":131},{},[132,134,139],{"type":62,"value":133},"Skip for ",{"type":57,"tag":74,"props":135,"children":136},{},[137],{"type":62,"value":138},"analytics-only",{"type":62,"value":140}," requests (summarize, chart, filter), fully\npre-specified math outside this flow, or explicit replayable\u002Fauditable path.",{"type":57,"tag":91,"props":142,"children":144},{"id":143},"sequence",[145],{"type":62,"value":146},"Sequence",{"type":57,"tag":65,"props":148,"children":149},{},[150,155,157,163],{"type":57,"tag":74,"props":151,"children":152},{},[153],{"type":62,"value":154},"Step 0 (NemoClaw — do not skip):",{"type":62,"value":156}," See ",{"type":57,"tag":82,"props":158,"children":160},{"className":159},[],[161],{"type":62,"value":162},"cuopt-sandbox",{"type":62,"value":164}," — probe → env →\nsmoke. No schedule\u002Fheuristic output before smoke passes.",{"type":57,"tag":166,"props":167,"children":168},"ol",{},[169,183,197,211],{"type":57,"tag":112,"props":170,"children":171},{},[172,181],{"type":57,"tag":74,"props":173,"children":174},{},[175],{"type":57,"tag":82,"props":176,"children":178},{"className":177},[],[179],{"type":62,"value":180},"optimization-intent-router",{"type":62,"value":182}," — optimization family (LP\u002FMILP\u002FQP\u002Frouting)",{"type":57,"tag":112,"props":184,"children":185},{},[186,195],{"type":57,"tag":74,"props":187,"children":188},{},[189],{"type":57,"tag":82,"props":190,"children":192},{"className":191},[],[193],{"type":62,"value":194},"optimization-mode-router",{"type":62,"value":196}," — only if replay\u002Faudit\u002Fexport signals",{"type":57,"tag":112,"props":198,"children":199},{},[200,209],{"type":57,"tag":74,"props":201,"children":202},{},[203],{"type":57,"tag":82,"props":204,"children":206},{"className":205},[],[207],{"type":62,"value":208},"tabular-optimization-ingestion",{"type":62,"value":210}," — table roles (interpretation only)",{"type":57,"tag":112,"props":212,"children":213},{},[214,223],{"type":57,"tag":74,"props":215,"children":216},{},[217],{"type":57,"tag":82,"props":218,"children":220},{"className":219},[],[221],{"type":62,"value":222},"cuopt-model-mapper",{"type":62,"value":224}," — clarify if needed, map to cuOpt, solve",{"type":57,"tag":65,"props":226,"children":227},{},[228],{"type":62,"value":229},"Handoffs after step 4:",{"type":57,"tag":108,"props":231,"children":232},{},[233,252],{"type":57,"tag":112,"props":234,"children":235},{},[236,238,244,246],{"type":62,"value":237},"LP \u002F MILP \u002F QP → ",{"type":57,"tag":82,"props":239,"children":241},{"className":240},[],[242],{"type":62,"value":243},"numerical-optimization-formulation",{"type":62,"value":245}," → ",{"type":57,"tag":82,"props":247,"children":249},{"className":248},[],[250],{"type":62,"value":251},"cuopt-numerical-optimization-api-python",{"type":57,"tag":112,"props":253,"children":254},{},[255,257,263,264],{"type":62,"value":256},"Routing → ",{"type":57,"tag":82,"props":258,"children":260},{"className":259},[],[261],{"type":62,"value":262},"routing-formulation",{"type":62,"value":245},{"type":57,"tag":82,"props":265,"children":267},{"className":266},[],[268],{"type":62,"value":269},"cuopt-routing-api-python",{"type":57,"tag":91,"props":271,"children":273},{"id":272},"guardrails",[274],{"type":62,"value":275},"Guardrails",{"type":57,"tag":108,"props":277,"children":278},{},[279,291,296,301],{"type":57,"tag":112,"props":280,"children":281},{},[282,284,289],{"type":62,"value":283},"First solver that emits assignments\u002Fschedules must be ",{"type":57,"tag":74,"props":285,"children":286},{},[287],{"type":62,"value":288},"cuOpt",{"type":62,"value":290}," after step 0",{"type":57,"tag":112,"props":292,"children":293},{},[294],{"type":62,"value":295},"Ingestion steps do not authorize heuristic or greedy stand-ins",{"type":57,"tag":112,"props":297,"children":298},{},[299],{"type":62,"value":300},"Do not skip intent classification; do not use cuOpt for pure analytics",{"type":57,"tag":112,"props":302,"children":303},{},[304],{"type":62,"value":305},"One focused clarification beats a long questionnaire",{"items":307,"total":391},[308,323,334,348,361,368,382],{"slug":309,"name":309,"fn":310,"description":311,"org":312,"tags":313,"stars":23,"repoUrl":24,"updatedAt":322},"cuopt-debugging","debug NVIDIA cuOpt optimization problems","Troubleshoot cuOpt LP\u002FMILP problems including errors, wrong results, infeasible solutions, performance issues, and status codes. Use when the user says something isn't working, gets unexpected results, or needs help diagnosing issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[314,317,320,321],{"name":315,"slug":316,"type":15},"Debugging","debugging",{"name":318,"slug":319,"type":15},"Mathematics","mathematics",{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},"2026-07-23T05:43:32.51428",{"slug":222,"name":222,"fn":324,"description":325,"org":326,"tags":327,"stars":23,"repoUrl":24,"updatedAt":333},"map optimization problems to cuOpt models","Map interpreted optimization problems into cuOpt-native models for the fast path with minimal clarifying questions.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[328,331,332],{"name":329,"slug":330,"type":15},"Data Modeling","data-modeling",{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},"2026-07-30T05:29:19.401656",{"slug":162,"name":162,"fn":335,"description":336,"org":337,"tags":338,"stars":23,"repoUrl":24,"updatedAt":347},"run cuOpt optimization in NemoClaw sandbox","Run cuOpt in the NemoClaw sandbox — probe\u002Fsmoke gates, prefer cancelable Python gRPC jobs, use legacy remote execution only when that API is unavailable, then vendored cuOpt skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[339,342,343,344],{"name":340,"slug":341,"type":15},"CLI","cli",{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},{"name":345,"slug":346,"type":15},"Simulation","simulation","2026-07-30T05:29:17.131364",{"slug":349,"name":349,"fn":350,"description":351,"org":352,"tags":353,"stars":23,"repoUrl":24,"updatedAt":360},"generic-max-supply","plan supply chain models with cuOpt","Multi-period supply chain planning model: data files, BOM structure, variable\u002Fconstraint reference for the max-supply base model.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[354,355,356,357],{"name":329,"slug":330,"type":15},{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},{"name":358,"slug":359,"type":15},"Supply Chain","supply-chain","2026-07-14T05:32:29.39577",{"slug":4,"name":4,"fn":5,"description":6,"org":362,"tags":363,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[364,365,366,367],{"name":17,"slug":18,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},{"slug":180,"name":180,"fn":369,"description":370,"org":371,"tags":372,"stars":23,"repoUrl":24,"updatedAt":381},"classify optimization and analytics requests","Classify whether a data-backed request is LP, MILP, QP, routing, or non-optimization analytics.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[373,376,379,380],{"name":374,"slug":375,"type":15},"Analytics","analytics",{"name":377,"slug":378,"type":15},"Data Analysis","data-analysis",{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},"2026-07-14T05:32:53.475243",{"slug":194,"name":194,"fn":383,"description":384,"org":385,"tags":386,"stars":23,"repoUrl":24,"updatedAt":390},"route optimization requests to cuOpt solvers","Choose fast direct-to-cuOpt solve versus replayable or auditable model artifact mode.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[387,388,389],{"name":17,"slug":18,"type":15},{"name":9,"slug":8,"type":15},{"name":20,"slug":21,"type":15},"2026-07-30T05:29:18.162351",8,{"items":393,"total":548},[394,412,430,441,453,465,478,492,505,516,530,539],{"slug":395,"name":395,"fn":396,"description":397,"org":398,"tags":399,"stars":409,"repoUrl":410,"updatedAt":411},"nemoclaw-user-guide","retrieve NemoClaw documentation and configuration","Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[400,403,406],{"name":401,"slug":402,"type":15},"Documentation","documentation",{"name":404,"slug":405,"type":15},"MCP","mcp",{"name":407,"slug":408,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":413,"name":413,"fn":414,"description":415,"org":416,"tags":417,"stars":427,"repoUrl":428,"updatedAt":429},"mcore-build-and-dependency","manage Megatron-LM development environments","Container-based dev environment setup and dependency management for Megatron-LM. Covers acquiring and launching the CI container, uv package management, and updating uv.lock.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[418,421,424],{"name":419,"slug":420,"type":15},"Containers","containers",{"name":422,"slug":423,"type":15},"Deployment","deployment",{"name":425,"slug":426,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":431,"name":431,"fn":432,"description":433,"org":434,"tags":435,"stars":427,"repoUrl":428,"updatedAt":440},"mcore-bump-base-image","update NVIDIA PyTorch base images","Bump the NVIDIA PyTorch base image (`nvcr.io\u002Fnvidia\u002Fpytorch:YY.MM-py3`) used by Megatron-LM CI. Covers the two pin sites (GitHub CI in `docker\u002F.ngc_version.dev` and GitLab CI in `.gitlab\u002Fstages\u002F01.build.yml`), the post-bump CI loop (re-run functional tests, refresh golden values, mark broken tests), and the gotchas that bit PRs",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[436,439],{"name":437,"slug":438,"type":15},"CI\u002FCD","ci-cd",{"name":422,"slug":423,"type":15},"2026-07-14T05:25:59.97109",{"slug":442,"name":442,"fn":443,"description":444,"org":445,"tags":446,"stars":427,"repoUrl":428,"updatedAt":452},"mcore-cicd","manage CI\u002FCD pipelines for Megatron-LM","CI\u002FCD reference for Megatron-LM. Covers CI pipeline structure, PR scope labels, triggering internal GitLab CI (which force-pushes the current branch to a pull-request\u002FBRANCH ref — always dry-run and verify the destination first; never run against shared or protected branches), and CI failure investigation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[447,448,449],{"name":437,"slug":438,"type":15},{"name":422,"slug":423,"type":15},{"name":450,"slug":451,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":454,"name":454,"fn":455,"description":456,"org":457,"tags":458,"stars":427,"repoUrl":428,"updatedAt":464},"mcore-create-issue","investigate CI failures and create issues","Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[459,460,461],{"name":315,"slug":316,"type":15},{"name":450,"slug":451,"type":15},{"name":462,"slug":463,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":466,"name":466,"fn":467,"description":468,"org":469,"tags":470,"stars":427,"repoUrl":428,"updatedAt":477},"mcore-linting-and-formatting","lint and format Megatron-LM code","Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[471,474],{"name":472,"slug":473,"type":15},"Best Practices","best-practices",{"name":475,"slug":476,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":479,"name":479,"fn":480,"description":481,"org":482,"tags":483,"stars":427,"repoUrl":428,"updatedAt":491},"mcore-migrate-gpt-to-hybrid","migrate Megatron-LM models to HybridModel","Migration guide for moving Megatron Core GPTModel checkpoints, model providers, training commands, and layer mappings to HybridModel.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[484,487,490],{"name":485,"slug":486,"type":15},"Machine Learning","machine-learning",{"name":488,"slug":489,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":493,"name":493,"fn":494,"description":495,"org":496,"tags":497,"stars":427,"repoUrl":428,"updatedAt":504},"mcore-onboard-gb200-1node-tests","onboard functional tests for GB200","Onboard 1-node GitHub MR functional tests for GB200 from existing mr-scoped 2-node tests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[498,501],{"name":499,"slug":500,"type":15},"QA","qa",{"name":502,"slug":503,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":506,"name":506,"fn":507,"description":508,"org":509,"tags":510,"stars":427,"repoUrl":428,"updatedAt":515},"mcore-run-on-slurm","launch distributed training jobs on SLURM","How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[511,512],{"name":422,"slug":423,"type":15},{"name":513,"slug":514,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":517,"name":517,"fn":518,"description":519,"org":520,"tags":521,"stars":427,"repoUrl":428,"updatedAt":529},"mcore-split-pr","split pull requests to reduce review load","Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[522,525,526],{"name":523,"slug":524,"type":15},"Code Review","code-review",{"name":450,"slug":451,"type":15},{"name":527,"slug":528,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":531,"name":531,"fn":532,"description":533,"org":534,"tags":535,"stars":427,"repoUrl":428,"updatedAt":538},"mcore-testing","run and manage Megatron-LM tests","Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[536,537],{"name":499,"slug":500,"type":15},{"name":502,"slug":503,"type":15},"2026-07-14T05:25:54.928983",{"slug":540,"name":540,"fn":541,"description":542,"org":543,"tags":544,"stars":427,"repoUrl":428,"updatedAt":547},"nightly-sync","manage nightly main-to-dev sync workflows","Domain knowledge for the nightly main-to-dev sync workflow. Covers merge strategy, CI architecture, failure investigation, and known issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[545,546],{"name":17,"slug":18,"type":15},{"name":437,"slug":438,"type":15},"2026-07-30T05:29:03.275638",496]