[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-skill-evolution":3,"mdc--rl575o-key":31,"related-repo-nvidia-skill-evolution":394,"related-org-nvidia-skill-evolution":504},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":20,"repoUrl":21,"updatedAt":22,"license":23,"forks":24,"topics":25,"repo":26,"sourceUrl":29,"mdContent":30},"skill-evolution","manage and evolve coding agent skills","Use before creating, editing, or deciding whether to update any AI coding agent skill in this repository, including corrections to existing skill behavior, references, or routing.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,17],{"name":13,"slug":14,"type":15},"Best Practices","best-practices","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Engineering","engineering",2491,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNeMo-Agent-Toolkit","2026-07-14T05:31:36.121876","Apache-2.0",704,[],{"repoUrl":21,"stars":20,"forks":24,"topics":27,"description":28},[],"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\u002Fskill-evolution","---\nname: skill-evolution\ndescription: Use before creating, editing, or deciding whether to update any AI coding agent skill in this repository, including corrections to existing skill behavior, references, or routing.\nauthor: NVIDIA Corporation and Affiliates\nlicense: Apache-2.0\n---\n\n# Skill Evolution\n\nUse this skill before creating, editing, or deciding whether to update any AI coding agent skill in this repository.\n\n## When to Update Skills\n\nUpdate a skill when any of the following happens:\n\n- A user corrects agent behavior in a way that should generalize.\n- A command fails and the recovery path should be remembered.\n- A reference link, command, component `_type`, or installation instruction is stale.\n- Task routing is too broad, too narrow, or points to the wrong skill.\n- A repeated instruction appears in multiple skills and should move to a shared entry point.\n- A skill contains too much detail and should move detail into focused references.\n\nDo not update skills for one-off user preferences, temporary local environment quirks, or speculative guidance that was not validated.\n\n## Update Workflow\n\n1. Finish the user's requested task first unless the skill update is the task.\n2. Identify the smallest skill that should change.\n3. Keep each `SKILL.md` concise and task-oriented.\n4. Move detailed examples or long reference material into that skill's `references\u002F` directory.\n5. Update `AGENTS.md` when a new skill is added or task routing changes.\n6. Run Markdown link checks through pre-commit on the changed skill files.\n\n## Naming and Layout\n\nUse the flat skills layout:\n\n```text\nskills\u002F\n  nat-installation\u002F\n    SKILL.md\n    references\u002F\n  nat-workflow-creation\u002F\n    SKILL.md\n    references\u002F\n```\n\nEach skill folder must contain one `SKILL.md` with frontmatter:\n\n```yaml\n---\nname: skill-name\ndescription: Use when ...\nauthor: NVIDIA Corporation and Affiliates\nlicense: Apache-2.0\n---\n```\n\nUse `author` and `license` frontmatter instead of long license headers in `SKILL.md` files.\n\nUse specific names that describe the task surface. Avoid catch-all folders that hide routing information.\n\n## Quality Bar\n\n- Skills should tell agents what to do, what to read next, and what to validate.\n- Prefer canonical repository docs over copied long-form explanations.\n- Keep cross-skill links relative and valid.\n- Use \"NVIDIA NeMo Agent Toolkit\" on first prose use, then \"NeMo Agent Toolkit\" or \"the toolkit\".\n- Use `nat` only for technical identifiers such as the CLI, package name, Python namespace, paths, and environment variables.\n",{"data":32,"body":34},{"name":4,"description":6,"author":33,"license":23},"NVIDIA Corporation and Affiliates",{"type":35,"children":36},"root",[37,45,51,58,63,107,112,118,176,182,187,199,211,316,342,347,353,388],{"type":38,"tag":39,"props":40,"children":41},"element","h1",{"id":4},[42],{"type":43,"value":44},"text","Skill Evolution",{"type":38,"tag":46,"props":47,"children":48},"p",{},[49],{"type":43,"value":50},"Use this skill before creating, editing, or deciding whether to update any AI coding agent skill in this repository.",{"type":38,"tag":52,"props":53,"children":55},"h2",{"id":54},"when-to-update-skills",[56],{"type":43,"value":57},"When to Update Skills",{"type":38,"tag":46,"props":59,"children":60},{},[61],{"type":43,"value":62},"Update a skill when any of the following happens:",{"type":38,"tag":64,"props":65,"children":66},"ul",{},[67,73,78,92,97,102],{"type":38,"tag":68,"props":69,"children":70},"li",{},[71],{"type":43,"value":72},"A user corrects agent behavior in a way that should generalize.",{"type":38,"tag":68,"props":74,"children":75},{},[76],{"type":43,"value":77},"A command fails and the recovery path should be remembered.",{"type":38,"tag":68,"props":79,"children":80},{},[81,83,90],{"type":43,"value":82},"A reference link, command, component ",{"type":38,"tag":84,"props":85,"children":87},"code",{"className":86},[],[88],{"type":43,"value":89},"_type",{"type":43,"value":91},", or installation instruction is stale.",{"type":38,"tag":68,"props":93,"children":94},{},[95],{"type":43,"value":96},"Task routing is too broad, too narrow, or points to the wrong skill.",{"type":38,"tag":68,"props":98,"children":99},{},[100],{"type":43,"value":101},"A repeated instruction appears in multiple skills and should move to a shared entry point.",{"type":38,"tag":68,"props":103,"children":104},{},[105],{"type":43,"value":106},"A skill contains too much detail and should move detail into focused references.",{"type":38,"tag":46,"props":108,"children":109},{},[110],{"type":43,"value":111},"Do not update skills for one-off user preferences, temporary local environment quirks, or speculative guidance that was not validated.",{"type":38,"tag":52,"props":113,"children":115},{"id":114},"update-workflow",[116],{"type":43,"value":117},"Update Workflow",{"type":38,"tag":119,"props":120,"children":121},"ol",{},[122,127,132,145,158,171],{"type":38,"tag":68,"props":123,"children":124},{},[125],{"type":43,"value":126},"Finish the user's requested task first unless the skill update is the task.",{"type":38,"tag":68,"props":128,"children":129},{},[130],{"type":43,"value":131},"Identify the smallest skill that should change.",{"type":38,"tag":68,"props":133,"children":134},{},[135,137,143],{"type":43,"value":136},"Keep each ",{"type":38,"tag":84,"props":138,"children":140},{"className":139},[],[141],{"type":43,"value":142},"SKILL.md",{"type":43,"value":144}," concise and task-oriented.",{"type":38,"tag":68,"props":146,"children":147},{},[148,150,156],{"type":43,"value":149},"Move detailed examples or long reference material into that skill's ",{"type":38,"tag":84,"props":151,"children":153},{"className":152},[],[154],{"type":43,"value":155},"references\u002F",{"type":43,"value":157}," directory.",{"type":38,"tag":68,"props":159,"children":160},{},[161,163,169],{"type":43,"value":162},"Update ",{"type":38,"tag":84,"props":164,"children":166},{"className":165},[],[167],{"type":43,"value":168},"AGENTS.md",{"type":43,"value":170}," when a new skill is added or task routing changes.",{"type":38,"tag":68,"props":172,"children":173},{},[174],{"type":43,"value":175},"Run Markdown link checks through pre-commit on the changed skill files.",{"type":38,"tag":52,"props":177,"children":179},{"id":178},"naming-and-layout",[180],{"type":43,"value":181},"Naming and Layout",{"type":38,"tag":46,"props":183,"children":184},{},[185],{"type":43,"value":186},"Use the flat skills layout:",{"type":38,"tag":188,"props":189,"children":194},"pre",{"className":190,"code":192,"language":43,"meta":193},[191],"language-text","skills\u002F\n  nat-installation\u002F\n    SKILL.md\n    references\u002F\n  nat-workflow-creation\u002F\n    SKILL.md\n    references\u002F\n","",[195],{"type":38,"tag":84,"props":196,"children":197},{"__ignoreMap":193},[198],{"type":43,"value":192},{"type":38,"tag":46,"props":200,"children":201},{},[202,204,209],{"type":43,"value":203},"Each skill folder must contain one ",{"type":38,"tag":84,"props":205,"children":207},{"className":206},[],[208],{"type":43,"value":142},{"type":43,"value":210}," with frontmatter:",{"type":38,"tag":188,"props":212,"children":216},{"className":213,"code":214,"language":215,"meta":193,"style":193},"language-yaml shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","---\nname: skill-name\ndescription: Use when ...\nauthor: NVIDIA Corporation and Affiliates\nlicense: Apache-2.0\n---\n","yaml",[217],{"type":38,"tag":84,"props":218,"children":219},{"__ignoreMap":193},[220,232,254,272,290,308],{"type":38,"tag":221,"props":222,"children":225},"span",{"class":223,"line":224},"line",1,[226],{"type":38,"tag":221,"props":227,"children":229},{"style":228},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[230],{"type":43,"value":231},"---\n",{"type":38,"tag":221,"props":233,"children":235},{"class":223,"line":234},2,[236,242,248],{"type":38,"tag":221,"props":237,"children":239},{"style":238},"--shiki-light:#E53935;--shiki-default:#F07178;--shiki-dark:#F07178",[240],{"type":43,"value":241},"name",{"type":38,"tag":221,"props":243,"children":245},{"style":244},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[246],{"type":43,"value":247},":",{"type":38,"tag":221,"props":249,"children":251},{"style":250},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[252],{"type":43,"value":253}," skill-name\n",{"type":38,"tag":221,"props":255,"children":257},{"class":223,"line":256},3,[258,263,267],{"type":38,"tag":221,"props":259,"children":260},{"style":238},[261],{"type":43,"value":262},"description",{"type":38,"tag":221,"props":264,"children":265},{"style":244},[266],{"type":43,"value":247},{"type":38,"tag":221,"props":268,"children":269},{"style":250},[270],{"type":43,"value":271}," Use when ...\n",{"type":38,"tag":221,"props":273,"children":275},{"class":223,"line":274},4,[276,281,285],{"type":38,"tag":221,"props":277,"children":278},{"style":238},[279],{"type":43,"value":280},"author",{"type":38,"tag":221,"props":282,"children":283},{"style":244},[284],{"type":43,"value":247},{"type":38,"tag":221,"props":286,"children":287},{"style":250},[288],{"type":43,"value":289}," NVIDIA Corporation and Affiliates\n",{"type":38,"tag":221,"props":291,"children":293},{"class":223,"line":292},5,[294,299,303],{"type":38,"tag":221,"props":295,"children":296},{"style":238},[297],{"type":43,"value":298},"license",{"type":38,"tag":221,"props":300,"children":301},{"style":244},[302],{"type":43,"value":247},{"type":38,"tag":221,"props":304,"children":305},{"style":250},[306],{"type":43,"value":307}," Apache-2.0\n",{"type":38,"tag":221,"props":309,"children":311},{"class":223,"line":310},6,[312],{"type":38,"tag":221,"props":313,"children":314},{"style":228},[315],{"type":43,"value":231},{"type":38,"tag":46,"props":317,"children":318},{},[319,321,326,328,333,335,340],{"type":43,"value":320},"Use ",{"type":38,"tag":84,"props":322,"children":324},{"className":323},[],[325],{"type":43,"value":280},{"type":43,"value":327}," and ",{"type":38,"tag":84,"props":329,"children":331},{"className":330},[],[332],{"type":43,"value":298},{"type":43,"value":334}," frontmatter instead of long license headers in ",{"type":38,"tag":84,"props":336,"children":338},{"className":337},[],[339],{"type":43,"value":142},{"type":43,"value":341}," files.",{"type":38,"tag":46,"props":343,"children":344},{},[345],{"type":43,"value":346},"Use specific names that describe the task surface. Avoid catch-all folders that hide routing information.",{"type":38,"tag":52,"props":348,"children":350},{"id":349},"quality-bar",[351],{"type":43,"value":352},"Quality Bar",{"type":38,"tag":64,"props":354,"children":355},{},[356,361,366,371,376],{"type":38,"tag":68,"props":357,"children":358},{},[359],{"type":43,"value":360},"Skills should tell agents what to do, what to read next, and what to validate.",{"type":38,"tag":68,"props":362,"children":363},{},[364],{"type":43,"value":365},"Prefer canonical repository docs over copied long-form explanations.",{"type":38,"tag":68,"props":367,"children":368},{},[369],{"type":43,"value":370},"Keep cross-skill links relative and valid.",{"type":38,"tag":68,"props":372,"children":373},{},[374],{"type":43,"value":375},"Use \"NVIDIA NeMo Agent Toolkit\" on first prose use, then \"NeMo Agent Toolkit\" or \"the toolkit\".",{"type":38,"tag":68,"props":377,"children":378},{},[379,380,386],{"type":43,"value":320},{"type":38,"tag":84,"props":381,"children":383},{"className":382},[],[384],{"type":43,"value":385},"nat",{"type":43,"value":387}," only for technical identifiers such as the CLI, package name, Python namespace, paths, and environment variables.",{"type":38,"tag":389,"props":390,"children":391},"style",{},[392],{"type":43,"value":393},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":395,"total":503},[396,410,427,439,454,469,483],{"slug":397,"name":397,"fn":398,"description":399,"org":400,"tags":401,"stars":20,"repoUrl":21,"updatedAt":409},"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},[402,405,406],{"name":403,"slug":404,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":407,"slug":408,"type":15},"Orchestration","orchestration","2026-07-14T05:31:37.436571",{"slug":411,"name":411,"fn":412,"description":413,"org":414,"tags":415,"stars":20,"repoUrl":21,"updatedAt":426},"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},[416,419,422,423],{"name":417,"slug":418,"type":15},"Datasets","datasets",{"name":420,"slug":421,"type":15},"Evals","evals",{"name":9,"slug":8,"type":15},{"name":424,"slug":425,"type":15},"Quality Assurance","quality-assurance","2026-07-14T05:31:45.038525",{"slug":428,"name":428,"fn":429,"description":430,"org":431,"tags":432,"stars":20,"repoUrl":21,"updatedAt":438},"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},[433,434,437],{"name":403,"slug":404,"type":15},{"name":435,"slug":436,"type":15},"Deployment","deployment",{"name":9,"slug":8,"type":15},"2026-07-14T05:31:34.833581",{"slug":440,"name":440,"fn":441,"description":442,"org":443,"tags":444,"stars":20,"repoUrl":21,"updatedAt":453},"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},[445,446,449,452],{"name":403,"slug":404,"type":15},{"name":447,"slug":448,"type":15},"FastAPI","fastapi",{"name":450,"slug":451,"type":15},"MCP","mcp",{"name":9,"slug":8,"type":15},"2026-07-14T05:31:33.571521",{"slug":455,"name":455,"fn":456,"description":457,"org":458,"tags":459,"stars":20,"repoUrl":21,"updatedAt":468},"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},[460,461,462,465],{"name":403,"slug":404,"type":15},{"name":9,"slug":8,"type":15},{"name":463,"slug":464,"type":15},"Optimization","optimization",{"name":466,"slug":467,"type":15},"Performance","performance","2026-07-14T05:31:32.290062",{"slug":470,"name":470,"fn":471,"description":472,"org":473,"tags":474,"stars":20,"repoUrl":21,"updatedAt":482},"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},[475,478,481],{"name":476,"slug":477,"type":15},"Debugging","debugging",{"name":479,"slug":480,"type":15},"Documentation","documentation",{"name":9,"slug":8,"type":15},"2026-07-14T05:31:42.534678",{"slug":484,"name":484,"fn":485,"description":486,"org":487,"tags":488,"stars":20,"repoUrl":21,"updatedAt":502},"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},[489,492,495,496,499],{"name":490,"slug":491,"type":15},"Langfuse","langfuse",{"name":493,"slug":494,"type":15},"LangSmith","langsmith",{"name":9,"slug":8,"type":15},{"name":497,"slug":498,"type":15},"Observability","observability",{"name":500,"slug":501,"type":15},"OpenTelemetry","opentelemetry","2026-07-14T05:31:41.276077",11,{"items":505,"total":654},[506,520,536,547,559,571,582,596,609,620,634,643],{"slug":507,"name":507,"fn":508,"description":509,"org":510,"tags":511,"stars":517,"repoUrl":518,"updatedAt":519},"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},[512,513,514],{"name":479,"slug":480,"type":15},{"name":450,"slug":451,"type":15},{"name":515,"slug":516,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":521,"name":521,"fn":522,"description":523,"org":524,"tags":525,"stars":533,"repoUrl":534,"updatedAt":535},"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},[526,529,530],{"name":527,"slug":528,"type":15},"Containers","containers",{"name":435,"slug":436,"type":15},{"name":531,"slug":532,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":537,"name":537,"fn":538,"description":539,"org":540,"tags":541,"stars":533,"repoUrl":534,"updatedAt":546},"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},[542,545],{"name":543,"slug":544,"type":15},"CI\u002FCD","ci-cd",{"name":435,"slug":436,"type":15},"2026-07-14T05:25:59.97109",{"slug":548,"name":548,"fn":549,"description":550,"org":551,"tags":552,"stars":533,"repoUrl":534,"updatedAt":558},"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},[553,554,555],{"name":543,"slug":544,"type":15},{"name":435,"slug":436,"type":15},{"name":556,"slug":557,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":560,"name":560,"fn":561,"description":562,"org":563,"tags":564,"stars":533,"repoUrl":534,"updatedAt":570},"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},[565,566,567],{"name":476,"slug":477,"type":15},{"name":556,"slug":557,"type":15},{"name":568,"slug":569,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":572,"name":572,"fn":573,"description":574,"org":575,"tags":576,"stars":533,"repoUrl":534,"updatedAt":581},"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},[577,578],{"name":13,"slug":14,"type":15},{"name":579,"slug":580,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":583,"name":583,"fn":584,"description":585,"org":586,"tags":587,"stars":533,"repoUrl":534,"updatedAt":595},"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},[588,591,594],{"name":589,"slug":590,"type":15},"Machine Learning","machine-learning",{"name":592,"slug":593,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":597,"name":597,"fn":598,"description":599,"org":600,"tags":601,"stars":533,"repoUrl":534,"updatedAt":608},"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},[602,605],{"name":603,"slug":604,"type":15},"QA","qa",{"name":606,"slug":607,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":610,"name":610,"fn":611,"description":612,"org":613,"tags":614,"stars":533,"repoUrl":534,"updatedAt":619},"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},[615,616],{"name":435,"slug":436,"type":15},{"name":617,"slug":618,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":621,"name":621,"fn":622,"description":623,"org":624,"tags":625,"stars":533,"repoUrl":534,"updatedAt":633},"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},[626,629,630],{"name":627,"slug":628,"type":15},"Code Review","code-review",{"name":556,"slug":557,"type":15},{"name":631,"slug":632,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":635,"name":635,"fn":636,"description":637,"org":638,"tags":639,"stars":533,"repoUrl":534,"updatedAt":642},"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},[640,641],{"name":603,"slug":604,"type":15},{"name":606,"slug":607,"type":15},"2026-07-14T05:25:54.928983",{"slug":644,"name":644,"fn":645,"description":646,"org":647,"tags":648,"stars":533,"repoUrl":534,"updatedAt":653},"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},[649,652],{"name":650,"slug":651,"type":15},"Automation","automation",{"name":543,"slug":544,"type":15},"2026-07-30T05:29:03.275638",496]