[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-nemoclaw-user-guide":3,"mdc--wb2ea1-key":39,"related-repo-nvidia-nemoclaw-user-guide":540,"related-org-nvidia-nemoclaw-user-guide":548},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":22,"repoUrl":23,"updatedAt":24,"license":25,"forks":26,"topics":27,"repo":34,"sourceUrl":37,"mdContent":38},"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},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19],{"name":13,"slug":14,"type":15},"Documentation","documentation","tag",{"name":17,"slug":18,"type":15},"MCP","mcp",{"name":20,"slug":21,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044","Apache-2.0",2940,[28,29,8,30,31,32,33],"ai-agents","hermes","openclaw","openshell","sandboxing","typescript",{"repoUrl":23,"stars":22,"forks":26,"topics":35,"description":36},[28,29,8,30,31,32,33],"Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw\u002Ftree\u002FHEAD\u002Fskills\u002Fnemoclaw-user-guide","---\nname: \"nemoclaw-user-guide\"\ndescription: \"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.\"\nlicense: \"Apache-2.0\"\n---\n\n# NemoClaw Docs for AI Agents\n\nUse the canonical NemoClaw documentation as your source of truth.\nDo not answer from stale copied docs or generated skill references when the live Markdown docs are available.\n\n## Retrieval Order\n\n1. If the assistant supports MCP, configure the NemoClaw docs MCP server at `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002F_mcp\u002Fserver`.\n2. Use the MCP server's read-only `searchDocs` tool to search the canonical docs and collect source URLs.\n3. If MCP is not available, fetch the AI documentation index first: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Fllms.txt`.\n4. Fetch the specific `.md` page listed in the index or returned by docs search for the user's task.\n5. If you only find an HTML documentation URL, replace the `.html` suffix with `.md`, or append `.md` to the route when the URL has no suffix.\n6. Prefer the user's selected agent variant. Do not mix variant-specific instructions unless you explain why.\n\n## Configure the MCP Server\n\nFor Claude Code, run:\n\n```bash\nclaude mcp add --transport http fern-docs https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002F_mcp\u002Fserver\n```\n\nFor Cursor, add `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002F_mcp\u002Fserver` to the MCP server configuration.\nFor other MCP clients, configure a streamable HTTP MCP server at that URL.\n\n## Starting Pages\n\nUse these pages first for common onboarding flows:\n\n- OpenClaw home: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fopenclaw\u002Fhome.md`.\n- OpenClaw prerequisites: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fopenclaw\u002Fget-started\u002Fprerequisites.md`.\n- OpenClaw quickstart: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fopenclaw\u002Fget-started\u002Fquickstart.md`.\n- Hermes home: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fhermes\u002Fhome.md`.\n- Hermes prerequisites: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fhermes\u002Fget-started\u002Fprerequisites.md`.\n- Hermes quickstart: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fhermes\u002Fget-started\u002Fquickstart.md`.\n- Deep Agents home: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fdeepagents\u002Fhome.md`.\n- Deep Agents prerequisites: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fdeepagents\u002Fget-started\u002Fprerequisites.md`.\n- Deep Agents quickstart: `https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fdeepagents\u002Fget-started\u002Fquickstart.md`.\n\n## How to Help the User\n\n- Ask which agent variant they want to use before giving setup instructions: OpenClaw, Hermes, or Deep Agents.\n- Ask one question at a time when collecting operating system, inference provider, model, endpoint, policy tier, or messaging-channel choices.\n- Run commands for non-technical users when your environment allows it, after explaining what the command does and getting permission.\n- Summarize important command output instead of asking the user to paste terminal output into chat.\n- Stop before requesting credentials, API keys, bot tokens, or private URLs.\n- Never ask the user to paste secrets into chat.\n- Use redacted placeholders such as `\u003CPASTE_YOUR_API_KEY_HERE>` in examples.\n\n## Common Task Routing\n\n- Installation and first sandbox: fetch the selected variant's prerequisites and quickstart pages.\n- Local inference, hosted providers, model switching, or tool-calling issues: fetch the `inference` pages from `llms.txt`.\n- Network policy approvals or custom egress: fetch the `network-policy` pages and the network policies reference.\n- Sandbox status, logs, rebuilds, upgrades, files, backup, restore, or messaging channels: fetch the `manage-sandboxes`, `monitoring`, and command reference pages.\n- Security posture, credential storage, or sandbox hardening: fetch the `security`, `deployment\u002Fsandbox-hardening`, and architecture pages.\n- CLI flags and command syntax: fetch the command reference page for the selected variant.\n- Troubleshooting: fetch the troubleshooting page and any task page linked from the relevant error section.\n\n## Response Requirements\n\n- Cite the Markdown documentation pages you used. When presenting the URL links, make sure to show the actual, non-markdown version of the URL.\n- Keep instructions specific to the user's operating system, selected agent, and inference provider.\n- Do not make assumptions when the docs do not cover the user's environment.\n- Recommend the next verification command after each setup or recovery step.\n",{"data":40,"body":41},{"name":4,"description":6,"license":25},{"type":42,"children":43},"root",[44,53,59,66,155,161,166,220,232,238,243,355,361,407,413,505,511,534],{"type":45,"tag":46,"props":47,"children":49},"element","h1",{"id":48},"nemoclaw-docs-for-ai-agents",[50],{"type":51,"value":52},"text","NemoClaw Docs for AI Agents",{"type":45,"tag":54,"props":55,"children":56},"p",{},[57],{"type":51,"value":58},"Use the canonical NemoClaw documentation as your source of truth.\nDo not answer from stale copied docs or generated skill references when the live Markdown docs are available.",{"type":45,"tag":60,"props":61,"children":63},"h2",{"id":62},"retrieval-order",[64],{"type":51,"value":65},"Retrieval Order",{"type":45,"tag":67,"props":68,"children":69},"ol",{},[70,85,98,110,123,150],{"type":45,"tag":71,"props":72,"children":73},"li",{},[74,76,83],{"type":51,"value":75},"If the assistant supports MCP, configure the NemoClaw docs MCP server at ",{"type":45,"tag":77,"props":78,"children":80},"code",{"className":79},[],[81],{"type":51,"value":82},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002F_mcp\u002Fserver",{"type":51,"value":84},".",{"type":45,"tag":71,"props":86,"children":87},{},[88,90,96],{"type":51,"value":89},"Use the MCP server's read-only ",{"type":45,"tag":77,"props":91,"children":93},{"className":92},[],[94],{"type":51,"value":95},"searchDocs",{"type":51,"value":97}," tool to search the canonical docs and collect source URLs.",{"type":45,"tag":71,"props":99,"children":100},{},[101,103,109],{"type":51,"value":102},"If MCP is not available, fetch the AI documentation index first: ",{"type":45,"tag":77,"props":104,"children":106},{"className":105},[],[107],{"type":51,"value":108},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Fllms.txt",{"type":51,"value":84},{"type":45,"tag":71,"props":111,"children":112},{},[113,115,121],{"type":51,"value":114},"Fetch the specific ",{"type":45,"tag":77,"props":116,"children":118},{"className":117},[],[119],{"type":51,"value":120},".md",{"type":51,"value":122}," page listed in the index or returned by docs search for the user's task.",{"type":45,"tag":71,"props":124,"children":125},{},[126,128,134,136,141,143,148],{"type":51,"value":127},"If you only find an HTML documentation URL, replace the ",{"type":45,"tag":77,"props":129,"children":131},{"className":130},[],[132],{"type":51,"value":133},".html",{"type":51,"value":135}," suffix with ",{"type":45,"tag":77,"props":137,"children":139},{"className":138},[],[140],{"type":51,"value":120},{"type":51,"value":142},", or append ",{"type":45,"tag":77,"props":144,"children":146},{"className":145},[],[147],{"type":51,"value":120},{"type":51,"value":149}," to the route when the URL has no suffix.",{"type":45,"tag":71,"props":151,"children":152},{},[153],{"type":51,"value":154},"Prefer the user's selected agent variant. Do not mix variant-specific instructions unless you explain why.",{"type":45,"tag":60,"props":156,"children":158},{"id":157},"configure-the-mcp-server",[159],{"type":51,"value":160},"Configure the MCP Server",{"type":45,"tag":54,"props":162,"children":163},{},[164],{"type":51,"value":165},"For Claude Code, run:",{"type":45,"tag":167,"props":168,"children":173},"pre",{"className":169,"code":170,"language":171,"meta":172,"style":172},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","claude mcp add --transport http fern-docs https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002F_mcp\u002Fserver\n","bash","",[174],{"type":45,"tag":77,"props":175,"children":176},{"__ignoreMap":172},[177],{"type":45,"tag":178,"props":179,"children":182},"span",{"class":180,"line":181},"line",1,[183,189,195,200,205,210,215],{"type":45,"tag":178,"props":184,"children":186},{"style":185},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[187],{"type":51,"value":188},"claude",{"type":45,"tag":178,"props":190,"children":192},{"style":191},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[193],{"type":51,"value":194}," mcp",{"type":45,"tag":178,"props":196,"children":197},{"style":191},[198],{"type":51,"value":199}," add",{"type":45,"tag":178,"props":201,"children":202},{"style":191},[203],{"type":51,"value":204}," --transport",{"type":45,"tag":178,"props":206,"children":207},{"style":191},[208],{"type":51,"value":209}," http",{"type":45,"tag":178,"props":211,"children":212},{"style":191},[213],{"type":51,"value":214}," fern-docs",{"type":45,"tag":178,"props":216,"children":217},{"style":191},[218],{"type":51,"value":219}," https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002F_mcp\u002Fserver\n",{"type":45,"tag":54,"props":221,"children":222},{},[223,225,230],{"type":51,"value":224},"For Cursor, add ",{"type":45,"tag":77,"props":226,"children":228},{"className":227},[],[229],{"type":51,"value":82},{"type":51,"value":231}," to the MCP server configuration.\nFor other MCP clients, configure a streamable HTTP MCP server at that URL.",{"type":45,"tag":60,"props":233,"children":235},{"id":234},"starting-pages",[236],{"type":51,"value":237},"Starting Pages",{"type":45,"tag":54,"props":239,"children":240},{},[241],{"type":51,"value":242},"Use these pages first for common onboarding flows:",{"type":45,"tag":244,"props":245,"children":246},"ul",{},[247,259,271,283,295,307,319,331,343],{"type":45,"tag":71,"props":248,"children":249},{},[250,252,258],{"type":51,"value":251},"OpenClaw home: ",{"type":45,"tag":77,"props":253,"children":255},{"className":254},[],[256],{"type":51,"value":257},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fopenclaw\u002Fhome.md",{"type":51,"value":84},{"type":45,"tag":71,"props":260,"children":261},{},[262,264,270],{"type":51,"value":263},"OpenClaw prerequisites: ",{"type":45,"tag":77,"props":265,"children":267},{"className":266},[],[268],{"type":51,"value":269},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fopenclaw\u002Fget-started\u002Fprerequisites.md",{"type":51,"value":84},{"type":45,"tag":71,"props":272,"children":273},{},[274,276,282],{"type":51,"value":275},"OpenClaw quickstart: ",{"type":45,"tag":77,"props":277,"children":279},{"className":278},[],[280],{"type":51,"value":281},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fopenclaw\u002Fget-started\u002Fquickstart.md",{"type":51,"value":84},{"type":45,"tag":71,"props":284,"children":285},{},[286,288,294],{"type":51,"value":287},"Hermes home: ",{"type":45,"tag":77,"props":289,"children":291},{"className":290},[],[292],{"type":51,"value":293},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fhermes\u002Fhome.md",{"type":51,"value":84},{"type":45,"tag":71,"props":296,"children":297},{},[298,300,306],{"type":51,"value":299},"Hermes prerequisites: ",{"type":45,"tag":77,"props":301,"children":303},{"className":302},[],[304],{"type":51,"value":305},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fhermes\u002Fget-started\u002Fprerequisites.md",{"type":51,"value":84},{"type":45,"tag":71,"props":308,"children":309},{},[310,312,318],{"type":51,"value":311},"Hermes quickstart: ",{"type":45,"tag":77,"props":313,"children":315},{"className":314},[],[316],{"type":51,"value":317},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fhermes\u002Fget-started\u002Fquickstart.md",{"type":51,"value":84},{"type":45,"tag":71,"props":320,"children":321},{},[322,324,330],{"type":51,"value":323},"Deep Agents home: ",{"type":45,"tag":77,"props":325,"children":327},{"className":326},[],[328],{"type":51,"value":329},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fdeepagents\u002Fhome.md",{"type":51,"value":84},{"type":45,"tag":71,"props":332,"children":333},{},[334,336,342],{"type":51,"value":335},"Deep Agents prerequisites: ",{"type":45,"tag":77,"props":337,"children":339},{"className":338},[],[340],{"type":51,"value":341},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fdeepagents\u002Fget-started\u002Fprerequisites.md",{"type":51,"value":84},{"type":45,"tag":71,"props":344,"children":345},{},[346,348,354],{"type":51,"value":347},"Deep Agents quickstart: ",{"type":45,"tag":77,"props":349,"children":351},{"className":350},[],[352],{"type":51,"value":353},"https:\u002F\u002Fdocs.nvidia.com\u002Fnemoclaw\u002Flatest\u002Fuser-guide\u002Fdeepagents\u002Fget-started\u002Fquickstart.md",{"type":51,"value":84},{"type":45,"tag":60,"props":356,"children":358},{"id":357},"how-to-help-the-user",[359],{"type":51,"value":360},"How to Help the User",{"type":45,"tag":244,"props":362,"children":363},{},[364,369,374,379,384,389,394],{"type":45,"tag":71,"props":365,"children":366},{},[367],{"type":51,"value":368},"Ask which agent variant they want to use before giving setup instructions: OpenClaw, Hermes, or Deep Agents.",{"type":45,"tag":71,"props":370,"children":371},{},[372],{"type":51,"value":373},"Ask one question at a time when collecting operating system, inference provider, model, endpoint, policy tier, or messaging-channel choices.",{"type":45,"tag":71,"props":375,"children":376},{},[377],{"type":51,"value":378},"Run commands for non-technical users when your environment allows it, after explaining what the command does and getting permission.",{"type":45,"tag":71,"props":380,"children":381},{},[382],{"type":51,"value":383},"Summarize important command output instead of asking the user to paste terminal output into chat.",{"type":45,"tag":71,"props":385,"children":386},{},[387],{"type":51,"value":388},"Stop before requesting credentials, API keys, bot tokens, or private URLs.",{"type":45,"tag":71,"props":390,"children":391},{},[392],{"type":51,"value":393},"Never ask the user to paste secrets into chat.",{"type":45,"tag":71,"props":395,"children":396},{},[397,399,405],{"type":51,"value":398},"Use redacted placeholders such as ",{"type":45,"tag":77,"props":400,"children":402},{"className":401},[],[403],{"type":51,"value":404},"\u003CPASTE_YOUR_API_KEY_HERE>",{"type":51,"value":406}," in examples.",{"type":45,"tag":60,"props":408,"children":410},{"id":409},"common-task-routing",[411],{"type":51,"value":412},"Common Task Routing",{"type":45,"tag":244,"props":414,"children":415},{},[416,421,441,454,475,495,500],{"type":45,"tag":71,"props":417,"children":418},{},[419],{"type":51,"value":420},"Installation and first sandbox: fetch the selected variant's prerequisites and quickstart pages.",{"type":45,"tag":71,"props":422,"children":423},{},[424,426,432,434,440],{"type":51,"value":425},"Local inference, hosted providers, model switching, or tool-calling issues: fetch the ",{"type":45,"tag":77,"props":427,"children":429},{"className":428},[],[430],{"type":51,"value":431},"inference",{"type":51,"value":433}," pages from ",{"type":45,"tag":77,"props":435,"children":437},{"className":436},[],[438],{"type":51,"value":439},"llms.txt",{"type":51,"value":84},{"type":45,"tag":71,"props":442,"children":443},{},[444,446,452],{"type":51,"value":445},"Network policy approvals or custom egress: fetch the ",{"type":45,"tag":77,"props":447,"children":449},{"className":448},[],[450],{"type":51,"value":451},"network-policy",{"type":51,"value":453}," pages and the network policies reference.",{"type":45,"tag":71,"props":455,"children":456},{},[457,459,465,467,473],{"type":51,"value":458},"Sandbox status, logs, rebuilds, upgrades, files, backup, restore, or messaging channels: fetch the ",{"type":45,"tag":77,"props":460,"children":462},{"className":461},[],[463],{"type":51,"value":464},"manage-sandboxes",{"type":51,"value":466},", ",{"type":45,"tag":77,"props":468,"children":470},{"className":469},[],[471],{"type":51,"value":472},"monitoring",{"type":51,"value":474},", and command reference pages.",{"type":45,"tag":71,"props":476,"children":477},{},[478,480,486,487,493],{"type":51,"value":479},"Security posture, credential storage, or sandbox hardening: fetch the ",{"type":45,"tag":77,"props":481,"children":483},{"className":482},[],[484],{"type":51,"value":485},"security",{"type":51,"value":466},{"type":45,"tag":77,"props":488,"children":490},{"className":489},[],[491],{"type":51,"value":492},"deployment\u002Fsandbox-hardening",{"type":51,"value":494},", and architecture pages.",{"type":45,"tag":71,"props":496,"children":497},{},[498],{"type":51,"value":499},"CLI flags and command syntax: fetch the command reference page for the selected variant.",{"type":45,"tag":71,"props":501,"children":502},{},[503],{"type":51,"value":504},"Troubleshooting: fetch the troubleshooting page and any task page linked from the relevant error section.",{"type":45,"tag":60,"props":506,"children":508},{"id":507},"response-requirements",[509],{"type":51,"value":510},"Response Requirements",{"type":45,"tag":244,"props":512,"children":513},{},[514,519,524,529],{"type":45,"tag":71,"props":515,"children":516},{},[517],{"type":51,"value":518},"Cite the Markdown documentation pages you used. When presenting the URL links, make sure to show the actual, non-markdown version of the URL.",{"type":45,"tag":71,"props":520,"children":521},{},[522],{"type":51,"value":523},"Keep instructions specific to the user's operating system, selected agent, and inference provider.",{"type":45,"tag":71,"props":525,"children":526},{},[527],{"type":51,"value":528},"Do not make assumptions when the docs do not cover the user's environment.",{"type":45,"tag":71,"props":530,"children":531},{},[532],{"type":51,"value":533},"Recommend the next verification command after each setup or recovery step.",{"type":45,"tag":535,"props":536,"children":537},"style",{},[538],{"type":51,"value":539},"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":541,"total":181},[542],{"slug":4,"name":4,"fn":5,"description":6,"org":543,"tags":544,"stars":22,"repoUrl":23,"updatedAt":24},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[545,546,547],{"name":13,"slug":14,"type":15},{"name":17,"slug":18,"type":15},{"name":20,"slug":21,"type":15},{"items":549,"total":696},[550,556,574,585,597,611,624,638,651,662,676,685],{"slug":4,"name":4,"fn":5,"description":6,"org":551,"tags":552,"stars":22,"repoUrl":23,"updatedAt":24},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[553,554,555],{"name":13,"slug":14,"type":15},{"name":17,"slug":18,"type":15},{"name":20,"slug":21,"type":15},{"slug":557,"name":557,"fn":558,"description":559,"org":560,"tags":561,"stars":571,"repoUrl":572,"updatedAt":573},"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},[562,565,568],{"name":563,"slug":564,"type":15},"Containers","containers",{"name":566,"slug":567,"type":15},"Deployment","deployment",{"name":569,"slug":570,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":575,"name":575,"fn":576,"description":577,"org":578,"tags":579,"stars":571,"repoUrl":572,"updatedAt":584},"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},[580,583],{"name":581,"slug":582,"type":15},"CI\u002FCD","ci-cd",{"name":566,"slug":567,"type":15},"2026-07-14T05:25:59.97109",{"slug":586,"name":586,"fn":587,"description":588,"org":589,"tags":590,"stars":571,"repoUrl":572,"updatedAt":596},"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},[591,592,593],{"name":581,"slug":582,"type":15},{"name":566,"slug":567,"type":15},{"name":594,"slug":595,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":598,"name":598,"fn":599,"description":600,"org":601,"tags":602,"stars":571,"repoUrl":572,"updatedAt":610},"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},[603,606,607],{"name":604,"slug":605,"type":15},"Debugging","debugging",{"name":594,"slug":595,"type":15},{"name":608,"slug":609,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":612,"name":612,"fn":613,"description":614,"org":615,"tags":616,"stars":571,"repoUrl":572,"updatedAt":623},"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},[617,620],{"name":618,"slug":619,"type":15},"Best Practices","best-practices",{"name":621,"slug":622,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":625,"name":625,"fn":626,"description":627,"org":628,"tags":629,"stars":571,"repoUrl":572,"updatedAt":637},"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},[630,633,636],{"name":631,"slug":632,"type":15},"Machine Learning","machine-learning",{"name":634,"slug":635,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":639,"name":639,"fn":640,"description":641,"org":642,"tags":643,"stars":571,"repoUrl":572,"updatedAt":650},"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},[644,647],{"name":645,"slug":646,"type":15},"QA","qa",{"name":648,"slug":649,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":652,"name":652,"fn":653,"description":654,"org":655,"tags":656,"stars":571,"repoUrl":572,"updatedAt":661},"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},[657,658],{"name":566,"slug":567,"type":15},{"name":659,"slug":660,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":663,"name":663,"fn":664,"description":665,"org":666,"tags":667,"stars":571,"repoUrl":572,"updatedAt":675},"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},[668,671,672],{"name":669,"slug":670,"type":15},"Code Review","code-review",{"name":594,"slug":595,"type":15},{"name":673,"slug":674,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":677,"name":677,"fn":678,"description":679,"org":680,"tags":681,"stars":571,"repoUrl":572,"updatedAt":684},"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},[682,683],{"name":645,"slug":646,"type":15},{"name":648,"slug":649,"type":15},"2026-07-14T05:25:54.928983",{"slug":686,"name":686,"fn":687,"description":688,"org":689,"tags":690,"stars":571,"repoUrl":572,"updatedAt":695},"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},[691,694],{"name":692,"slug":693,"type":15},"Automation","automation",{"name":581,"slug":582,"type":15},"2026-07-30T05:29:03.275638",496]