[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-ragskill":3,"mdc--td0zan-key":45,"related-org-nvidia-ragskill":786,"related-repo-nvidia-ragskill":942},{"slug":4,"name":5,"fn":6,"description":7,"org":8,"tags":12,"stars":24,"repoUrl":25,"updatedAt":26,"license":27,"forks":28,"topics":29,"repo":40,"sourceUrl":43,"mdContent":44},"ragskill","rag_skill","retrieve simulator documentation and examples","Retrieve information from simulator manual and example DATA files. Use when answering keyword format questions, syntax queries, or when looking up official documentation and working examples. Essential for understanding keyword definitions, parameter tables, and concrete usage patterns.",{"slug":9,"name":10,"logoUrl":11,"githubOrg":10},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[13,17,20,21],{"name":14,"slug":15,"type":16},"Research","research","tag",{"name":18,"slug":19,"type":16},"Documentation","documentation",{"name":10,"slug":9,"type":16},{"name":22,"slug":23,"type":16},"Simulation","simulation",4115,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FGenerativeAIExamples","2026-07-14T05:32:15.32946","Apache-2.0",1093,[30,31,32,33,34,35,36,37,38,39],"gpu-acceleration","large-language-models","llm","llm-inference","microservice","nemo","rag","retrieval-augmented-generation","tensorrt","triton-inference-server",{"repoUrl":25,"stars":24,"forks":28,"topics":41,"description":42},[30,31,32,33,34,35,36,37,38,39],"Generative AI reference workflows optimized for accelerated infrastructure and microservice architecture.","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FGenerativeAIExamples\u002Ftree\u002FHEAD\u002Findustries\u002Fenergy\u002Fsimulation-workflow-agent\u002Fsim_agent\u002Fsrc\u002Fsimulator_agent\u002Fskills\u002Frag_skill","---\nname: rag_skill\ndescription: Retrieve information from simulator manual and example DATA files. Use when answering keyword format questions, syntax queries, or when looking up official documentation and working examples. Essential for understanding keyword definitions, parameter tables, and concrete usage patterns.\nlicense: Apache-2.0\nmetadata:\n  author: sim-agent\n  version: \"1.0\"\n  category: simulator\ncompatibility: Requires vector database with simulator manual and examples, embeddings model, and Python environment with langchain packages\n---\n\n# RAG (Retrieval-Augmented Generation) Skill\n\nThis skill provides retrieval tools for accessing simulator documentation and example files through vector search.\n\n## Overview\n\nThe RAG skill enables agents to:\n\n- **Retrieve** official simulator manual documentation for keyword definitions and syntax\n- **Retrieve** example DATA files and case studies for concrete usage patterns\n- **Answer** keyword format questions with authoritative sources\n- **Provide** working examples to illustrate keyword usage\n\n## Tools\n\n### simulator_manual\n\nRetrieves information from the simulator manual and official documentation using semantic search.\n\n**Usage:**\n```python\nsimulator_manual(query: str) -> str\n```\n\n**Parameters:**\n- `query`: Natural language query about keywords, syntax, or documentation (e.g., \"COMPDAT keyword format\", \"WELSPECS syntax and fields\")\n\n**Example:**\n```\nsimulator_manual(\"What is the COMPDAT keyword format?\")\n```\n\n**Returns:** Retrieved documentation snippets from the simulator manual with source citations.\n\n**When to use:** \n- First step in keyword Q&A flow (TOOL_DECISION_TREE.md Section 2.5)\n- In scenario test chain after `parse_simulation_input_file` (Section 2.4) to get keyword context\n- When user asks about keyword definitions, syntax, or parameter tables\n\n### simulator_examples\n\nRetrieves example DATA files and case studies using semantic search.\n\n**Usage:**\n```python\nsimulator_examples(query: str) -> str\n```\n\n**Parameters:**\n- `query`: Natural language query about keyword examples or usage patterns (e.g., \"COMPDAT keyword format\", \"WCONINJE injection rate examples\")\n\n**Example:**\n```\nsimulator_examples(\"COMPDAT keyword format\")\n```\n\n**Returns:** Retrieved example DATA file snippets showing concrete keyword usage.\n\n**When to use:**\n- After `simulator_manual` when manual lacks format details or examples (Section 2.5)\n- In scenario test chain after `simulator_manual` to get example context for modifications (Section 2.4)\n- When user needs working examples to understand keyword syntax\n\n## Workflow Integration\n\nThis skill integrates with the Simulator Agent's decision tree (TOOL_DECISION_TREE.md):\n\n1. **Keyword Q&A Flow (Section 2.5):**\n   ```\n   simulator_manual → (simulator_examples) → answer\n   ```\n\n2. **Scenario Test Chain (Section 2.4):**\n   ```\n   parse_simulation_input_file → simulator_manual (inferred keyword) → simulator_examples (same keyword) → modify_simulation_input_file → run_and_heal\n   ```\n\n3. **Keyword Chain (Section 3.4):**\n   ```\n   simulator_manual → simulator_examples → synthesize format + example → final answer\n   ```\n\n## Implementation Details\n\nTools are implemented as LangChain retriever tools with:\n- **Vector store**: Milvus containing embedded simulator manual and examples\n- **Embeddings**: NVIDIA embedding API for semantic search\n- **Top-k retrieval**: Configurable number of relevant chunks returned (default: 10, reranked to 5)\n- **Source citation**: Metadata includes source file paths and section references\n\n### extract_keyword (internal helper)\n\n`scripts\u002Fextract_keyword.py` provides RAG + LLM keyword extraction (e.g. \"plot field oil\" → FOPT). Used by plot_skill validators and other skills that need to infer keywords from natural language.\n\n```python\nfrom simulator_agent.skills.rag_skill.scripts.extract_keyword import extract_keyword\n\nkw = extract_keyword(\"plot field cumulative oil production\", intent=\"summary_metric\")  # -> \"FOPT\"\n```\n\n## Best Practices\n\n1. **Always cite sources**: Include manual section names and example file paths in responses\n2. **Use simulator_manual first**: It's the authoritative source for syntax and fields\n3. **Use simulator_examples for illustration**: Examples show concrete usage but should not override manual definitions\n4. **Handle conflicts**: If examples conflict with manual, prefer manual and explain the conflict\n\n## References\n\n- [Tool Decision Tree](references\u002FTOOL_DECISION_TREE.md) - Routing logic\n- [OPM Flow Manual](references\u002FOPM_MANUAL.md) - Manual structure and indexing\n\n## Configuration\n\nRAG tools use Milvus collections created by `.\u002Fscripts\u002Fsetup.sh --full`:\n\n| Tool name           | Milvus collection         | Ingested by                    |\n|---------------------|---------------------------|--------------------------------|\n| `simulator_manual`  | `docs`                    | `ingest_papers.sh`             |\n| `simulator_examples`| `simulator_input_examples`| `ingest_opm_examples.py`       |\n\nEnvironment variables:\n- `MILVUS_URI`: Milvus endpoint (default: `http:\u002F\u002Flocalhost:19530`; Docker: `http:\u002F\u002Fstandalone:19530`)\n- `NVIDIA_API_KEY`: Required for embeddings, reranker, and LLM\n\n",{"data":46,"body":52},{"name":5,"description":7,"license":27,"metadata":47,"compatibility":51},{"author":48,"version":49,"category":50},"sim-agent","1.0","simulator","Requires vector database with simulator manual and examples, embeddings model, and Python environment with langchain packages",{"type":53,"children":54},"root",[55,64,70,77,82,127,133,139,144,152,174,182,196,204,214,224,232,258,263,268,275,289,296,309,316,325,334,341,372,378,383,438,444,449,492,498,509,543,549,592,598,624,630,643,734,739,780],{"type":56,"tag":57,"props":58,"children":60},"element","h1",{"id":59},"rag-retrieval-augmented-generation-skill",[61],{"type":62,"value":63},"text","RAG (Retrieval-Augmented Generation) Skill",{"type":56,"tag":65,"props":66,"children":67},"p",{},[68],{"type":62,"value":69},"This skill provides retrieval tools for accessing simulator documentation and example files through vector search.",{"type":56,"tag":71,"props":72,"children":74},"h2",{"id":73},"overview",[75],{"type":62,"value":76},"Overview",{"type":56,"tag":65,"props":78,"children":79},{},[80],{"type":62,"value":81},"The RAG skill enables agents to:",{"type":56,"tag":83,"props":84,"children":85},"ul",{},[86,98,107,117],{"type":56,"tag":87,"props":88,"children":89},"li",{},[90,96],{"type":56,"tag":91,"props":92,"children":93},"strong",{},[94],{"type":62,"value":95},"Retrieve",{"type":62,"value":97}," official simulator manual documentation for keyword definitions and syntax",{"type":56,"tag":87,"props":99,"children":100},{},[101,105],{"type":56,"tag":91,"props":102,"children":103},{},[104],{"type":62,"value":95},{"type":62,"value":106}," example DATA files and case studies for concrete usage patterns",{"type":56,"tag":87,"props":108,"children":109},{},[110,115],{"type":56,"tag":91,"props":111,"children":112},{},[113],{"type":62,"value":114},"Answer",{"type":62,"value":116}," keyword format questions with authoritative sources",{"type":56,"tag":87,"props":118,"children":119},{},[120,125],{"type":56,"tag":91,"props":121,"children":122},{},[123],{"type":62,"value":124},"Provide",{"type":62,"value":126}," working examples to illustrate keyword usage",{"type":56,"tag":71,"props":128,"children":130},{"id":129},"tools",[131],{"type":62,"value":132},"Tools",{"type":56,"tag":134,"props":135,"children":137},"h3",{"id":136},"simulator_manual",[138],{"type":62,"value":136},{"type":56,"tag":65,"props":140,"children":141},{},[142],{"type":62,"value":143},"Retrieves information from the simulator manual and official documentation using semantic search.",{"type":56,"tag":65,"props":145,"children":146},{},[147],{"type":56,"tag":91,"props":148,"children":149},{},[150],{"type":62,"value":151},"Usage:",{"type":56,"tag":153,"props":154,"children":159},"pre",{"className":155,"code":156,"language":157,"meta":158,"style":158},"language-python shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","simulator_manual(query: str) -> str\n","python","",[160],{"type":56,"tag":161,"props":162,"children":163},"code",{"__ignoreMap":158},[164],{"type":56,"tag":165,"props":166,"children":169},"span",{"class":167,"line":168},"line",1,[170],{"type":56,"tag":165,"props":171,"children":172},{},[173],{"type":62,"value":156},{"type":56,"tag":65,"props":175,"children":176},{},[177],{"type":56,"tag":91,"props":178,"children":179},{},[180],{"type":62,"value":181},"Parameters:",{"type":56,"tag":83,"props":183,"children":184},{},[185],{"type":56,"tag":87,"props":186,"children":187},{},[188,194],{"type":56,"tag":161,"props":189,"children":191},{"className":190},[],[192],{"type":62,"value":193},"query",{"type":62,"value":195},": Natural language query about keywords, syntax, or documentation (e.g., \"COMPDAT keyword format\", \"WELSPECS syntax and fields\")",{"type":56,"tag":65,"props":197,"children":198},{},[199],{"type":56,"tag":91,"props":200,"children":201},{},[202],{"type":62,"value":203},"Example:",{"type":56,"tag":153,"props":205,"children":209},{"className":206,"code":208,"language":62},[207],"language-text","simulator_manual(\"What is the COMPDAT keyword format?\")\n",[210],{"type":56,"tag":161,"props":211,"children":212},{"__ignoreMap":158},[213],{"type":62,"value":208},{"type":56,"tag":65,"props":215,"children":216},{},[217,222],{"type":56,"tag":91,"props":218,"children":219},{},[220],{"type":62,"value":221},"Returns:",{"type":62,"value":223}," Retrieved documentation snippets from the simulator manual with source citations.",{"type":56,"tag":65,"props":225,"children":226},{},[227],{"type":56,"tag":91,"props":228,"children":229},{},[230],{"type":62,"value":231},"When to use:",{"type":56,"tag":83,"props":233,"children":234},{},[235,240,253],{"type":56,"tag":87,"props":236,"children":237},{},[238],{"type":62,"value":239},"First step in keyword Q&A flow (TOOL_DECISION_TREE.md Section 2.5)",{"type":56,"tag":87,"props":241,"children":242},{},[243,245,251],{"type":62,"value":244},"In scenario test chain after ",{"type":56,"tag":161,"props":246,"children":248},{"className":247},[],[249],{"type":62,"value":250},"parse_simulation_input_file",{"type":62,"value":252}," (Section 2.4) to get keyword context",{"type":56,"tag":87,"props":254,"children":255},{},[256],{"type":62,"value":257},"When user asks about keyword definitions, syntax, or parameter tables",{"type":56,"tag":134,"props":259,"children":261},{"id":260},"simulator_examples",[262],{"type":62,"value":260},{"type":56,"tag":65,"props":264,"children":265},{},[266],{"type":62,"value":267},"Retrieves example DATA files and case studies using semantic search.",{"type":56,"tag":65,"props":269,"children":270},{},[271],{"type":56,"tag":91,"props":272,"children":273},{},[274],{"type":62,"value":151},{"type":56,"tag":153,"props":276,"children":278},{"className":155,"code":277,"language":157,"meta":158,"style":158},"simulator_examples(query: str) -> str\n",[279],{"type":56,"tag":161,"props":280,"children":281},{"__ignoreMap":158},[282],{"type":56,"tag":165,"props":283,"children":284},{"class":167,"line":168},[285],{"type":56,"tag":165,"props":286,"children":287},{},[288],{"type":62,"value":277},{"type":56,"tag":65,"props":290,"children":291},{},[292],{"type":56,"tag":91,"props":293,"children":294},{},[295],{"type":62,"value":181},{"type":56,"tag":83,"props":297,"children":298},{},[299],{"type":56,"tag":87,"props":300,"children":301},{},[302,307],{"type":56,"tag":161,"props":303,"children":305},{"className":304},[],[306],{"type":62,"value":193},{"type":62,"value":308},": Natural language query about keyword examples or usage patterns (e.g., \"COMPDAT keyword format\", \"WCONINJE injection rate examples\")",{"type":56,"tag":65,"props":310,"children":311},{},[312],{"type":56,"tag":91,"props":313,"children":314},{},[315],{"type":62,"value":203},{"type":56,"tag":153,"props":317,"children":320},{"className":318,"code":319,"language":62},[207],"simulator_examples(\"COMPDAT keyword format\")\n",[321],{"type":56,"tag":161,"props":322,"children":323},{"__ignoreMap":158},[324],{"type":62,"value":319},{"type":56,"tag":65,"props":326,"children":327},{},[328,332],{"type":56,"tag":91,"props":329,"children":330},{},[331],{"type":62,"value":221},{"type":62,"value":333}," Retrieved example DATA file snippets showing concrete keyword usage.",{"type":56,"tag":65,"props":335,"children":336},{},[337],{"type":56,"tag":91,"props":338,"children":339},{},[340],{"type":62,"value":231},{"type":56,"tag":83,"props":342,"children":343},{},[344,356,367],{"type":56,"tag":87,"props":345,"children":346},{},[347,349,354],{"type":62,"value":348},"After ",{"type":56,"tag":161,"props":350,"children":352},{"className":351},[],[353],{"type":62,"value":136},{"type":62,"value":355}," when manual lacks format details or examples (Section 2.5)",{"type":56,"tag":87,"props":357,"children":358},{},[359,360,365],{"type":62,"value":244},{"type":56,"tag":161,"props":361,"children":363},{"className":362},[],[364],{"type":62,"value":136},{"type":62,"value":366}," to get example context for modifications (Section 2.4)",{"type":56,"tag":87,"props":368,"children":369},{},[370],{"type":62,"value":371},"When user needs working examples to understand keyword syntax",{"type":56,"tag":71,"props":373,"children":375},{"id":374},"workflow-integration",[376],{"type":62,"value":377},"Workflow Integration",{"type":56,"tag":65,"props":379,"children":380},{},[381],{"type":62,"value":382},"This skill integrates with the Simulator Agent's decision tree (TOOL_DECISION_TREE.md):",{"type":56,"tag":384,"props":385,"children":386},"ol",{},[387,404,421],{"type":56,"tag":87,"props":388,"children":389},{},[390,395],{"type":56,"tag":91,"props":391,"children":392},{},[393],{"type":62,"value":394},"Keyword Q&A Flow (Section 2.5):",{"type":56,"tag":153,"props":396,"children":399},{"className":397,"code":398,"language":62},[207],"simulator_manual → (simulator_examples) → answer\n",[400],{"type":56,"tag":161,"props":401,"children":402},{"__ignoreMap":158},[403],{"type":62,"value":398},{"type":56,"tag":87,"props":405,"children":406},{},[407,412],{"type":56,"tag":91,"props":408,"children":409},{},[410],{"type":62,"value":411},"Scenario Test Chain (Section 2.4):",{"type":56,"tag":153,"props":413,"children":416},{"className":414,"code":415,"language":62},[207],"parse_simulation_input_file → simulator_manual (inferred keyword) → simulator_examples (same keyword) → modify_simulation_input_file → run_and_heal\n",[417],{"type":56,"tag":161,"props":418,"children":419},{"__ignoreMap":158},[420],{"type":62,"value":415},{"type":56,"tag":87,"props":422,"children":423},{},[424,429],{"type":56,"tag":91,"props":425,"children":426},{},[427],{"type":62,"value":428},"Keyword Chain (Section 3.4):",{"type":56,"tag":153,"props":430,"children":433},{"className":431,"code":432,"language":62},[207],"simulator_manual → simulator_examples → synthesize format + example → final answer\n",[434],{"type":56,"tag":161,"props":435,"children":436},{"__ignoreMap":158},[437],{"type":62,"value":432},{"type":56,"tag":71,"props":439,"children":441},{"id":440},"implementation-details",[442],{"type":62,"value":443},"Implementation Details",{"type":56,"tag":65,"props":445,"children":446},{},[447],{"type":62,"value":448},"Tools are implemented as LangChain retriever tools with:",{"type":56,"tag":83,"props":450,"children":451},{},[452,462,472,482],{"type":56,"tag":87,"props":453,"children":454},{},[455,460],{"type":56,"tag":91,"props":456,"children":457},{},[458],{"type":62,"value":459},"Vector store",{"type":62,"value":461},": Milvus containing embedded simulator manual and examples",{"type":56,"tag":87,"props":463,"children":464},{},[465,470],{"type":56,"tag":91,"props":466,"children":467},{},[468],{"type":62,"value":469},"Embeddings",{"type":62,"value":471},": NVIDIA embedding API for semantic search",{"type":56,"tag":87,"props":473,"children":474},{},[475,480],{"type":56,"tag":91,"props":476,"children":477},{},[478],{"type":62,"value":479},"Top-k retrieval",{"type":62,"value":481},": Configurable number of relevant chunks returned (default: 10, reranked to 5)",{"type":56,"tag":87,"props":483,"children":484},{},[485,490],{"type":56,"tag":91,"props":486,"children":487},{},[488],{"type":62,"value":489},"Source citation",{"type":62,"value":491},": Metadata includes source file paths and section references",{"type":56,"tag":134,"props":493,"children":495},{"id":494},"extract_keyword-internal-helper",[496],{"type":62,"value":497},"extract_keyword (internal helper)",{"type":56,"tag":65,"props":499,"children":500},{},[501,507],{"type":56,"tag":161,"props":502,"children":504},{"className":503},[],[505],{"type":62,"value":506},"scripts\u002Fextract_keyword.py",{"type":62,"value":508}," provides RAG + LLM keyword extraction (e.g. \"plot field oil\" → FOPT). Used by plot_skill validators and other skills that need to infer keywords from natural language.",{"type":56,"tag":153,"props":510,"children":512},{"className":155,"code":511,"language":157,"meta":158,"style":158},"from simulator_agent.skills.rag_skill.scripts.extract_keyword import extract_keyword\n\nkw = extract_keyword(\"plot field cumulative oil production\", intent=\"summary_metric\")  # -> \"FOPT\"\n",[513],{"type":56,"tag":161,"props":514,"children":515},{"__ignoreMap":158},[516,524,534],{"type":56,"tag":165,"props":517,"children":518},{"class":167,"line":168},[519],{"type":56,"tag":165,"props":520,"children":521},{},[522],{"type":62,"value":523},"from simulator_agent.skills.rag_skill.scripts.extract_keyword import extract_keyword\n",{"type":56,"tag":165,"props":525,"children":527},{"class":167,"line":526},2,[528],{"type":56,"tag":165,"props":529,"children":531},{"emptyLinePlaceholder":530},true,[532],{"type":62,"value":533},"\n",{"type":56,"tag":165,"props":535,"children":537},{"class":167,"line":536},3,[538],{"type":56,"tag":165,"props":539,"children":540},{},[541],{"type":62,"value":542},"kw = extract_keyword(\"plot field cumulative oil production\", intent=\"summary_metric\")  # -> \"FOPT\"\n",{"type":56,"tag":71,"props":544,"children":546},{"id":545},"best-practices",[547],{"type":62,"value":548},"Best Practices",{"type":56,"tag":384,"props":550,"children":551},{},[552,562,572,582],{"type":56,"tag":87,"props":553,"children":554},{},[555,560],{"type":56,"tag":91,"props":556,"children":557},{},[558],{"type":62,"value":559},"Always cite sources",{"type":62,"value":561},": Include manual section names and example file paths in responses",{"type":56,"tag":87,"props":563,"children":564},{},[565,570],{"type":56,"tag":91,"props":566,"children":567},{},[568],{"type":62,"value":569},"Use simulator_manual first",{"type":62,"value":571},": It's the authoritative source for syntax and fields",{"type":56,"tag":87,"props":573,"children":574},{},[575,580],{"type":56,"tag":91,"props":576,"children":577},{},[578],{"type":62,"value":579},"Use simulator_examples for illustration",{"type":62,"value":581},": Examples show concrete usage but should not override manual definitions",{"type":56,"tag":87,"props":583,"children":584},{},[585,590],{"type":56,"tag":91,"props":586,"children":587},{},[588],{"type":62,"value":589},"Handle conflicts",{"type":62,"value":591},": If examples conflict with manual, prefer manual and explain the conflict",{"type":56,"tag":71,"props":593,"children":595},{"id":594},"references",[596],{"type":62,"value":597},"References",{"type":56,"tag":83,"props":599,"children":600},{},[601,613],{"type":56,"tag":87,"props":602,"children":603},{},[604,611],{"type":56,"tag":605,"props":606,"children":608},"a",{"href":607},"references\u002FTOOL_DECISION_TREE.md",[609],{"type":62,"value":610},"Tool Decision Tree",{"type":62,"value":612}," - Routing logic",{"type":56,"tag":87,"props":614,"children":615},{},[616,622],{"type":56,"tag":605,"props":617,"children":619},{"href":618},"references\u002FOPM_MANUAL.md",[620],{"type":62,"value":621},"OPM Flow Manual",{"type":62,"value":623}," - Manual structure and indexing",{"type":56,"tag":71,"props":625,"children":627},{"id":626},"configuration",[628],{"type":62,"value":629},"Configuration",{"type":56,"tag":65,"props":631,"children":632},{},[633,635,641],{"type":62,"value":634},"RAG tools use Milvus collections created by ",{"type":56,"tag":161,"props":636,"children":638},{"className":637},[],[639],{"type":62,"value":640},".\u002Fscripts\u002Fsetup.sh --full",{"type":62,"value":642},":",{"type":56,"tag":644,"props":645,"children":646},"table",{},[647,671],{"type":56,"tag":648,"props":649,"children":650},"thead",{},[651],{"type":56,"tag":652,"props":653,"children":654},"tr",{},[655,661,666],{"type":56,"tag":656,"props":657,"children":658},"th",{},[659],{"type":62,"value":660},"Tool name",{"type":56,"tag":656,"props":662,"children":663},{},[664],{"type":62,"value":665},"Milvus collection",{"type":56,"tag":656,"props":667,"children":668},{},[669],{"type":62,"value":670},"Ingested by",{"type":56,"tag":672,"props":673,"children":674},"tbody",{},[675,705],{"type":56,"tag":652,"props":676,"children":677},{},[678,687,696],{"type":56,"tag":679,"props":680,"children":681},"td",{},[682],{"type":56,"tag":161,"props":683,"children":685},{"className":684},[],[686],{"type":62,"value":136},{"type":56,"tag":679,"props":688,"children":689},{},[690],{"type":56,"tag":161,"props":691,"children":693},{"className":692},[],[694],{"type":62,"value":695},"docs",{"type":56,"tag":679,"props":697,"children":698},{},[699],{"type":56,"tag":161,"props":700,"children":702},{"className":701},[],[703],{"type":62,"value":704},"ingest_papers.sh",{"type":56,"tag":652,"props":706,"children":707},{},[708,716,725],{"type":56,"tag":679,"props":709,"children":710},{},[711],{"type":56,"tag":161,"props":712,"children":714},{"className":713},[],[715],{"type":62,"value":260},{"type":56,"tag":679,"props":717,"children":718},{},[719],{"type":56,"tag":161,"props":720,"children":722},{"className":721},[],[723],{"type":62,"value":724},"simulator_input_examples",{"type":56,"tag":679,"props":726,"children":727},{},[728],{"type":56,"tag":161,"props":729,"children":731},{"className":730},[],[732],{"type":62,"value":733},"ingest_opm_examples.py",{"type":56,"tag":65,"props":735,"children":736},{},[737],{"type":62,"value":738},"Environment variables:",{"type":56,"tag":83,"props":740,"children":741},{},[742,769],{"type":56,"tag":87,"props":743,"children":744},{},[745,751,753,759,761,767],{"type":56,"tag":161,"props":746,"children":748},{"className":747},[],[749],{"type":62,"value":750},"MILVUS_URI",{"type":62,"value":752},": Milvus endpoint (default: ",{"type":56,"tag":161,"props":754,"children":756},{"className":755},[],[757],{"type":62,"value":758},"http:\u002F\u002Flocalhost:19530",{"type":62,"value":760},"; Docker: ",{"type":56,"tag":161,"props":762,"children":764},{"className":763},[],[765],{"type":62,"value":766},"http:\u002F\u002Fstandalone:19530",{"type":62,"value":768},")",{"type":56,"tag":87,"props":770,"children":771},{},[772,778],{"type":56,"tag":161,"props":773,"children":775},{"className":774},[],[776],{"type":62,"value":777},"NVIDIA_API_KEY",{"type":62,"value":779},": Required for embeddings, reranker, and LLM",{"type":56,"tag":781,"props":782,"children":783},"style",{},[784],{"type":62,"value":785},"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":787,"total":941},[788,804,821,832,844,858,869,883,896,907,921,930],{"slug":789,"name":789,"fn":790,"description":791,"org":792,"tags":793,"stars":801,"repoUrl":802,"updatedAt":803},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[794,795,798],{"name":18,"slug":19,"type":16},{"name":796,"slug":797,"type":16},"MCP","mcp",{"name":799,"slug":800,"type":16},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":805,"name":805,"fn":806,"description":807,"org":808,"tags":809,"stars":818,"repoUrl":819,"updatedAt":820},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[810,813,816],{"name":811,"slug":812,"type":16},"Containers","containers",{"name":814,"slug":815,"type":16},"Deployment","deployment",{"name":817,"slug":157,"type":16},"Python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":822,"name":822,"fn":823,"description":824,"org":825,"tags":826,"stars":818,"repoUrl":819,"updatedAt":831},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[827,830],{"name":828,"slug":829,"type":16},"CI\u002FCD","ci-cd",{"name":814,"slug":815,"type":16},"2026-07-14T05:25:59.97109",{"slug":833,"name":833,"fn":834,"description":835,"org":836,"tags":837,"stars":818,"repoUrl":819,"updatedAt":843},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[838,839,840],{"name":828,"slug":829,"type":16},{"name":814,"slug":815,"type":16},{"name":841,"slug":842,"type":16},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":845,"name":845,"fn":846,"description":847,"org":848,"tags":849,"stars":818,"repoUrl":819,"updatedAt":857},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[850,853,854],{"name":851,"slug":852,"type":16},"Debugging","debugging",{"name":841,"slug":842,"type":16},{"name":855,"slug":856,"type":16},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":859,"name":859,"fn":860,"description":861,"org":862,"tags":863,"stars":818,"repoUrl":819,"updatedAt":868},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[864,865],{"name":548,"slug":545,"type":16},{"name":866,"slug":867,"type":16},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":870,"name":870,"fn":871,"description":872,"org":873,"tags":874,"stars":818,"repoUrl":819,"updatedAt":882},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[875,878,881],{"name":876,"slug":877,"type":16},"Machine Learning","machine-learning",{"name":879,"slug":880,"type":16},"Migration","migration",{"name":10,"slug":9,"type":16},"2026-07-17T06:07:11.777011",{"slug":884,"name":884,"fn":885,"description":886,"org":887,"tags":888,"stars":818,"repoUrl":819,"updatedAt":895},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[889,892],{"name":890,"slug":891,"type":16},"QA","qa",{"name":893,"slug":894,"type":16},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":897,"name":897,"fn":898,"description":899,"org":900,"tags":901,"stars":818,"repoUrl":819,"updatedAt":906},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[902,903],{"name":814,"slug":815,"type":16},{"name":904,"slug":905,"type":16},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":908,"name":908,"fn":909,"description":910,"org":911,"tags":912,"stars":818,"repoUrl":819,"updatedAt":920},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[913,916,917],{"name":914,"slug":915,"type":16},"Code Review","code-review",{"name":841,"slug":842,"type":16},{"name":918,"slug":919,"type":16},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":922,"name":922,"fn":923,"description":924,"org":925,"tags":926,"stars":818,"repoUrl":819,"updatedAt":929},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[927,928],{"name":890,"slug":891,"type":16},{"name":893,"slug":894,"type":16},"2026-07-14T05:25:54.928983",{"slug":931,"name":931,"fn":932,"description":933,"org":934,"tags":935,"stars":818,"repoUrl":819,"updatedAt":940},"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":9,"name":10,"logoUrl":11,"githubOrg":10},[936,939],{"name":937,"slug":938,"type":16},"Automation","automation",{"name":828,"slug":829,"type":16},"2026-07-30T05:29:03.275638",496,{"items":943,"total":1011},[944,958,974,981,997],{"slug":945,"name":946,"fn":947,"description":948,"org":949,"tags":950,"stars":24,"repoUrl":25,"updatedAt":957},"inputfileskill","input_file_skill","manage simulator input files","Parse, modify, validate, and patch simulator input files. Use when working with reservoir simulation input files, testing scenarios, or validating simulation configurations. This implementation supports reference format (.DATA); other simulators use different extensions (e.g., .afi, .DAT). Supports natural language modifications, keyword patching, and syntax validation.",{"slug":9,"name":10,"logoUrl":11,"githubOrg":10},[951,952,953,954],{"name":629,"slug":626,"type":16},{"name":10,"slug":9,"type":16},{"name":22,"slug":23,"type":16},{"name":955,"slug":956,"type":16},"Validation","validation","2026-07-14T05:32:16.602719",{"slug":959,"name":960,"fn":961,"description":962,"org":963,"tags":964,"stars":24,"repoUrl":25,"updatedAt":973},"plotskill","plot_skill","visualize simulation summary metrics","Plot and compare simulation summary metrics. Use when visualizing time-series results, comparing multiple cases, or analyzing production performance. Supports single and multi-metric plots, case comparisons, and automatic metric keyword resolution.",{"slug":9,"name":10,"logoUrl":11,"githubOrg":10},[965,968,971,972],{"name":966,"slug":967,"type":16},"Charts","charts",{"name":969,"slug":970,"type":16},"Data Visualization","data-visualization",{"name":10,"slug":9,"type":16},{"name":22,"slug":23,"type":16},"2026-07-14T05:32:19.171135",{"slug":4,"name":5,"fn":6,"description":7,"org":975,"tags":976,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":9,"name":10,"logoUrl":11,"githubOrg":10},[977,978,979,980],{"name":18,"slug":19,"type":16},{"name":10,"slug":9,"type":16},{"name":14,"slug":15,"type":16},{"name":22,"slug":23,"type":16},{"slug":982,"name":983,"fn":984,"description":985,"org":986,"tags":987,"stars":24,"repoUrl":25,"updatedAt":996},"resultsskill","results_skill","analyze simulation binary output files","Read and analyze simulation binary output files. Use when extracting summary data, grid properties, or running flow diagnostics (time-of-flight, tracer, allocation, F-Phi, Lorenz) from completed simulations.",{"slug":9,"name":10,"logoUrl":11,"githubOrg":10},[988,991,992,995],{"name":989,"slug":990,"type":16},"Data Analysis","data-analysis",{"name":10,"slug":9,"type":16},{"name":993,"slug":994,"type":16},"Performance","performance",{"name":22,"slug":23,"type":16},"2026-07-14T05:32:14.078951",{"slug":998,"name":999,"fn":1000,"description":1001,"org":1002,"tags":1003,"stars":24,"repoUrl":25,"updatedAt":1010},"simulationskill","simulation_skill","run and monitor simulations","Run, monitor, and control simulations. Use when executing simulations, checking progress, or stopping running simulations. Supports foreground and background execution, progress monitoring, and process management.",{"slug":9,"name":10,"logoUrl":11,"githubOrg":10},[1004,1005,1008,1009],{"name":937,"slug":938,"type":16},{"name":1006,"slug":1007,"type":16},"Monitoring","monitoring",{"name":10,"slug":9,"type":16},{"name":22,"slug":23,"type":16},"2026-07-14T05:32:17.914931",5]