[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-nat-evaluation":3,"mdc-ygke1r-key":34,"related-repo-nvidia-nat-evaluation":174,"related-org-nvidia-nat-evaluation":274},{"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":29,"sourceUrl":32,"mdContent":33},"nat-evaluation","run evaluations for NeMo Agent Toolkit","Use when designing, configuring, running, or troubleshooting NeMo Agent Toolkit evaluations, datasets, evaluator selection, ATIF surfaces, quality gates, custom evaluators, and `nat eval`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,20],{"name":13,"slug":14,"type":15},"Datasets","datasets","tag",{"name":17,"slug":18,"type":15},"Evals","evals",{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},"Quality Assurance","quality-assurance",2491,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNeMo-Agent-Toolkit","2026-07-14T05:31:45.038525","Apache-2.0",704,[],{"repoUrl":24,"stars":23,"forks":27,"topics":30,"description":31},[],"The NVIDIA NeMo Agent toolkit is an open-source library for efficiently connecting and optimizing teams of AI agents.","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNeMo-Agent-Toolkit\u002Ftree\u002FHEAD\u002Fskills\u002Fnat-evaluation","---\nname: nat-evaluation\ndescription: Use when designing, configuring, running, or troubleshooting NeMo Agent Toolkit evaluations, datasets, evaluator selection, ATIF surfaces, quality gates, custom evaluators, and `nat eval`.\nauthor: NVIDIA Corporation and Affiliates\nlicense: Apache-2.0\n---\n\n# NeMo Agent Toolkit Evaluation\n\nUse this skill for measuring agent quality and behavior.\n\n## Workflow\n\n1. Decide the evaluation surface and output format.\n2. Decompose quality goals into separate evaluators.\n3. Choose built-in evaluators before writing custom evaluators.\n4. Keep datasets small and explicit for local validation.\n5. Run `nat eval` and inspect generated artifacts.\n\n## References\n\n- `references\u002Foperating-mode.md`\n- `references\u002Fmethodology.md`\n- `references\u002Fagent-eval-framework.md`\n- `references\u002Fevaluation-surfaces.md`\n- `references\u002Fevaluation-contract.md`\n- `references\u002Fevaluators\u002F`\n- `references\u002Fcode-patterns.md`\n",{"data":35,"body":37},{"name":4,"description":6,"author":36,"license":26},"NVIDIA Corporation and Affiliates",{"type":38,"children":39},"root",[40,49,55,62,101,107],{"type":41,"tag":42,"props":43,"children":45},"element","h1",{"id":44},"nemo-agent-toolkit-evaluation",[46],{"type":47,"value":48},"text","NeMo Agent Toolkit Evaluation",{"type":41,"tag":50,"props":51,"children":52},"p",{},[53],{"type":47,"value":54},"Use this skill for measuring agent quality and behavior.",{"type":41,"tag":56,"props":57,"children":59},"h2",{"id":58},"workflow",[60],{"type":47,"value":61},"Workflow",{"type":41,"tag":63,"props":64,"children":65},"ol",{},[66,72,77,82,87],{"type":41,"tag":67,"props":68,"children":69},"li",{},[70],{"type":47,"value":71},"Decide the evaluation surface and output format.",{"type":41,"tag":67,"props":73,"children":74},{},[75],{"type":47,"value":76},"Decompose quality goals into separate evaluators.",{"type":41,"tag":67,"props":78,"children":79},{},[80],{"type":47,"value":81},"Choose built-in evaluators before writing custom evaluators.",{"type":41,"tag":67,"props":83,"children":84},{},[85],{"type":47,"value":86},"Keep datasets small and explicit for local validation.",{"type":41,"tag":67,"props":88,"children":89},{},[90,92,99],{"type":47,"value":91},"Run ",{"type":41,"tag":93,"props":94,"children":96},"code",{"className":95},[],[97],{"type":47,"value":98},"nat eval",{"type":47,"value":100}," and inspect generated artifacts.",{"type":41,"tag":56,"props":102,"children":104},{"id":103},"references",[105],{"type":47,"value":106},"References",{"type":41,"tag":108,"props":109,"children":110},"ul",{},[111,120,129,138,147,156,165],{"type":41,"tag":67,"props":112,"children":113},{},[114],{"type":41,"tag":93,"props":115,"children":117},{"className":116},[],[118],{"type":47,"value":119},"references\u002Foperating-mode.md",{"type":41,"tag":67,"props":121,"children":122},{},[123],{"type":41,"tag":93,"props":124,"children":126},{"className":125},[],[127],{"type":47,"value":128},"references\u002Fmethodology.md",{"type":41,"tag":67,"props":130,"children":131},{},[132],{"type":41,"tag":93,"props":133,"children":135},{"className":134},[],[136],{"type":47,"value":137},"references\u002Fagent-eval-framework.md",{"type":41,"tag":67,"props":139,"children":140},{},[141],{"type":41,"tag":93,"props":142,"children":144},{"className":143},[],[145],{"type":47,"value":146},"references\u002Fevaluation-surfaces.md",{"type":41,"tag":67,"props":148,"children":149},{},[150],{"type":41,"tag":93,"props":151,"children":153},{"className":152},[],[154],{"type":47,"value":155},"references\u002Fevaluation-contract.md",{"type":41,"tag":67,"props":157,"children":158},{},[159],{"type":41,"tag":93,"props":160,"children":162},{"className":161},[],[163],{"type":47,"value":164},"references\u002Fevaluators\u002F",{"type":41,"tag":67,"props":166,"children":167},{},[168],{"type":41,"tag":93,"props":169,"children":171},{"className":170},[],[172],{"type":47,"value":173},"references\u002Fcode-patterns.md",{"items":175,"total":273},[176,190,197,209,224,239,253],{"slug":177,"name":177,"fn":178,"description":179,"org":180,"tags":181,"stars":23,"repoUrl":24,"updatedAt":189},"nat-agent-configuration","configure and compose NeMo Agent Toolkit agents","Use when selecting, configuring, composing, or troubleshooting NeMo Agent Toolkit agents and control-flow components, including ReAct, tool-calling, ReWOO, reasoning, router, sequential, parallel, and sub-agent patterns.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[182,185,186],{"name":183,"slug":184,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":187,"slug":188,"type":15},"Orchestration","orchestration","2026-07-14T05:31:37.436571",{"slug":4,"name":4,"fn":5,"description":6,"org":191,"tags":192,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[193,194,195,196],{"name":13,"slug":14,"type":15},{"name":17,"slug":18,"type":15},{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},{"slug":198,"name":198,"fn":199,"description":200,"org":201,"tags":202,"stars":23,"repoUrl":24,"updatedAt":208},"nat-installation","install and configure NeMo Agent Toolkit","Use when installing or configuring NVIDIA NeMo Agent Toolkit, verifying the `nat` CLI, setting up optional extras, or creating a first hello-world workflow.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[203,204,207],{"name":183,"slug":184,"type":15},{"name":205,"slug":206,"type":15},"Deployment","deployment",{"name":9,"slug":8,"type":15},"2026-07-14T05:31:34.833581",{"slug":210,"name":210,"fn":211,"description":212,"org":213,"tags":214,"stars":23,"repoUrl":24,"updatedAt":223},"nat-mcp-and-serving","serve and configure NeMo agent workflows","Use when serving NeMo Agent Toolkit workflows, exposing workflows through FastAPI, configuring MCP clients or servers, or troubleshooting transport and server setup.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[215,216,219,222],{"name":183,"slug":184,"type":15},{"name":217,"slug":218,"type":15},"FastAPI","fastapi",{"name":220,"slug":221,"type":15},"MCP","mcp",{"name":9,"slug":8,"type":15},"2026-07-14T05:31:33.571521",{"slug":225,"name":225,"fn":226,"description":227,"org":228,"tags":229,"stars":23,"repoUrl":24,"updatedAt":238},"nat-optimization","optimize agent workflows with NeMo","Use when configuring or running NeMo Agent Toolkit optimization with `nat optimize`, including Optuna parameter tuning, prompt evolution, optimizer sizing, output interpretation, and optimizer datasets.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[230,231,232,235],{"name":183,"slug":184,"type":15},{"name":9,"slug":8,"type":15},{"name":233,"slug":234,"type":15},"Optimization","optimization",{"name":236,"slug":237,"type":15},"Performance","performance","2026-07-14T05:31:32.290062",{"slug":240,"name":240,"fn":241,"description":242,"org":243,"tags":244,"stars":23,"repoUrl":24,"updatedAt":252},"nat-path-checks","fix NeMo Agent Toolkit documentation path errors","Use when fixing NeMo Agent Toolkit documentation path-check failures, especially failed `ci\u002Fscripts\u002Fpath_checks.py` output, slash-delimited text mistaken for paths, relative path references, Markdown code escaping, and path-check allowlist decisions.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[245,248,251],{"name":246,"slug":247,"type":15},"Debugging","debugging",{"name":249,"slug":250,"type":15},"Documentation","documentation",{"name":9,"slug":8,"type":15},"2026-07-14T05:31:42.534678",{"slug":254,"name":254,"fn":255,"description":256,"org":257,"tags":258,"stars":23,"repoUrl":24,"updatedAt":272},"nat-telemetry","configure telemetry for NeMo Agent Toolkit","Use when adding, configuring, or troubleshooting NeMo Agent Toolkit logging, tracing, telemetry exporters, OpenTelemetry, Langfuse, LangSmith, Weave, Phoenix, profiling, or observability provider integrations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[259,262,265,266,269],{"name":260,"slug":261,"type":15},"Langfuse","langfuse",{"name":263,"slug":264,"type":15},"LangSmith","langsmith",{"name":9,"slug":8,"type":15},{"name":267,"slug":268,"type":15},"Observability","observability",{"name":270,"slug":271,"type":15},"OpenTelemetry","opentelemetry","2026-07-14T05:31:41.276077",11,{"items":275,"total":426},[276,290,306,317,329,341,354,368,381,392,406,415],{"slug":277,"name":277,"fn":278,"description":279,"org":280,"tags":281,"stars":287,"repoUrl":288,"updatedAt":289},"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},[282,283,284],{"name":249,"slug":250,"type":15},{"name":220,"slug":221,"type":15},{"name":285,"slug":286,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":291,"name":291,"fn":292,"description":293,"org":294,"tags":295,"stars":303,"repoUrl":304,"updatedAt":305},"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},[296,299,300],{"name":297,"slug":298,"type":15},"Containers","containers",{"name":205,"slug":206,"type":15},{"name":301,"slug":302,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":307,"name":307,"fn":308,"description":309,"org":310,"tags":311,"stars":303,"repoUrl":304,"updatedAt":316},"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},[312,315],{"name":313,"slug":314,"type":15},"CI\u002FCD","ci-cd",{"name":205,"slug":206,"type":15},"2026-07-14T05:25:59.97109",{"slug":318,"name":318,"fn":319,"description":320,"org":321,"tags":322,"stars":303,"repoUrl":304,"updatedAt":328},"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},[323,324,325],{"name":313,"slug":314,"type":15},{"name":205,"slug":206,"type":15},{"name":326,"slug":327,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":330,"name":330,"fn":331,"description":332,"org":333,"tags":334,"stars":303,"repoUrl":304,"updatedAt":340},"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},[335,336,337],{"name":246,"slug":247,"type":15},{"name":326,"slug":327,"type":15},{"name":338,"slug":339,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":342,"name":342,"fn":343,"description":344,"org":345,"tags":346,"stars":303,"repoUrl":304,"updatedAt":353},"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},[347,350],{"name":348,"slug":349,"type":15},"Best Practices","best-practices",{"name":351,"slug":352,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":355,"name":355,"fn":356,"description":357,"org":358,"tags":359,"stars":303,"repoUrl":304,"updatedAt":367},"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},[360,363,366],{"name":361,"slug":362,"type":15},"Machine Learning","machine-learning",{"name":364,"slug":365,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":369,"name":369,"fn":370,"description":371,"org":372,"tags":373,"stars":303,"repoUrl":304,"updatedAt":380},"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},[374,377],{"name":375,"slug":376,"type":15},"QA","qa",{"name":378,"slug":379,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":382,"name":382,"fn":383,"description":384,"org":385,"tags":386,"stars":303,"repoUrl":304,"updatedAt":391},"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},[387,388],{"name":205,"slug":206,"type":15},{"name":389,"slug":390,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":393,"name":393,"fn":394,"description":395,"org":396,"tags":397,"stars":303,"repoUrl":304,"updatedAt":405},"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},[398,401,402],{"name":399,"slug":400,"type":15},"Code Review","code-review",{"name":326,"slug":327,"type":15},{"name":403,"slug":404,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":407,"name":407,"fn":408,"description":409,"org":410,"tags":411,"stars":303,"repoUrl":304,"updatedAt":414},"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},[412,413],{"name":375,"slug":376,"type":15},{"name":378,"slug":379,"type":15},"2026-07-14T05:25:54.928983",{"slug":416,"name":416,"fn":417,"description":418,"org":419,"tags":420,"stars":303,"repoUrl":304,"updatedAt":425},"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},[421,424],{"name":422,"slug":423,"type":15},"Automation","automation",{"name":313,"slug":314,"type":15},"2026-07-30T05:29:03.275638",496]