[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-nvidia-kaggle-skill":3,"mdc--d13va0-key":31,"related-repo-nvidia-nvidia-kaggle-skill":1504,"related-org-nvidia-nvidia-kaggle-skill":1512},{"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},"nvidia-kaggle-skill","manage Kaggle competition research and submissions","Use for Kaggle competition overview fetches, writeups, discussion\u002Fkernel research, submissions, and dataset uploads. Not for unrelated ML code.",{"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},"Research","research","tag",{"name":17,"slug":18,"type":15},"Machine Learning","machine-learning",{"name":9,"slug":8,"type":15},253,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnvidia-kaggle","2026-07-14T05:36:16.524177","MIT",22,[],{"repoUrl":21,"stars":20,"forks":24,"topics":27,"description":28},[],"NVIDIA Kaggle Plugin gives agents end-to-end Kaggle competition workflows through a single skill, nvidia-kaggle-skill. It can gather competition context, study public writeups and notebooks, reproduce kernels locally, submit to competitions, and manage Ka","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnvidia-kaggle\u002Ftree\u002FHEAD\u002Fskills\u002Fnvidia-kaggle-skill","---\nname: nvidia-kaggle-skill\ndescription: \"Use for Kaggle competition overview fetches, writeups, discussion\u002Fkernel research, submissions, and dataset uploads. Not for unrelated ML code.\"\nlicense: MIT\npermissions:\n  - \"shell: run the bundled scripts\u002F, the Kaggle CLI (kaggle), basic file utilities (mkdir, cd, cp), and install runtime Python packages\"\n  - \"network: HTTPS to Kaggle APIs (kaggle.com, api.kaggle.com) and PyPI\"\n  - \"env: read environment variables, including KAGGLE_API_TOKEN, and load a project .env file\"\n  - \"file_read: read the project .env, inputs under the skill workspace, and user-specified paths\"\n  - \"file_write: write reports, caches, and downloads under the skill workspace and user-specified paths\"\nmetadata:\n  short-description: \"Kaggle competition workflows\"\n  author: \"nvidia-kaggle maintainers\"\n  tags:\n    - kaggle\n    - competition\n    - data-science\n    - kernels\n---\n\n# NVIDIA Kaggle Skill\n\n## Purpose\n\nUse this skill for Kaggle competition work: context gathering, writeups, discussions, kernels, local reproduction, submission, and dataset upload.\n\nDo not use it for unrelated ML training, generic notebook editing, general data analysis, or non-Kaggle dataset management unless the user explicitly ties the task to Kaggle.\n\n## Inputs\n\n| Input | Required | Description |\n|---|---|---|\n| Kaggle slug, URL, writeup URL, kernel ref, or local folder | Depends on task | Primary target for the requested Kaggle action. |\n| `KAGGLE_API_TOKEN` | Required for API\u002FCLI-backed workflows | KGAT token string for Kaggle API, CLI, and SDK calls. |\n| Disk space | Required for kernel setup | Must fit input datasets, competition data, models, and extracted archives. |\n\n## Prerequisites\n\n- Run commands from this skill directory unless a referenced workflow says otherwise.\n- Install only the runtime packages needed for the requested workflow.\n- Set `KAGGLE_API_TOKEN` before API, CLI, kernel, discussion, dataset, or submission workflows.\n- Confirm local disk space before downloading competition data, kernel inputs, or extracted archives.\n- Require explicit user confirmation before sensitive, externally visible actions: competition submissions (each can consume a daily slot), dataset uploads, and creating a public dataset. Treat `KAGGLE_API_TOKEN` as a secret — never print, log, or echo it.\n\n## Runtime Dependencies\n\nInstall only the packages needed for the requested task into the current environment, then run scripts with `python`.\n\nKaggle API, CLI, kernels, discussions, datasets, competition pages, and writeups:\n\n```bash\nif command -v uv >\u002Fdev\u002Fnull 2>&1; then\n  uv pip install httpx kaggle kagglesdk nbformat pydantic python-dotenv rich\nelse\n  python -m pip install httpx kaggle kagglesdk nbformat pydantic python-dotenv rich\nfi\n```\n\nFor API\u002FCLI tasks, verify credentials before calling Kaggle:\n\n```bash\n: \"${KAGGLE_API_TOKEN:?ERROR: KAGGLE_API_TOKEN environment variable is not set}\"\n```\n\n## Workflows\n\nUse this workflow catalog to choose the right path. Run the direct script commands for quick tasks. For workflows that point to another markdown file, read that file only when the request needs that workflow.\n\nPrefer the runtime's `run_script` helper when it exists, for example `run_script(\"scripts\u002Ffetch_competition_info.py\", args=[\"titanic\"])`. Otherwise run the equivalent `python .\u002Fscripts\u002F\u003Cscript>.py ...` command from this skill directory.\n\n### Competition Details\n\nUse this when the user asks to retrieve or summarize a Kaggle competition overview, rules, evaluation, timeline, or dataset description.\n\nFetch overview:\n\n```bash\npython .\u002Fscripts\u002Ffetch_competition_info.py \u003Ccompetition-slug-or-url>\n```\n\nFetch dataset description:\n\n```bash\npython .\u002Fscripts\u002Ffetch_dataset_info.py \u003Ccompetition-slug-or-url>\n```\n\nThe scripts accept a bare competition slug or `https:\u002F\u002Fwww.kaggle.com\u002Fcompetitions\u002F\u003Cslug>` URL and extract the slug automatically. Convert output to markdown when the user asks for saved documentation, using `{slug}_competition_overview.md` and `{slug}_dataset_description.md` in the current working directory.\n\n### Research Brief\n\nUse this when the user asks you to **research a competition and write a strategy\nbrief** in natural terms (e.g. \"research this competition and brief me, with\nlinks and a few charts\"). You chain the skill's individual research workflows\nyourself, write your own analysis\u002Fplotting code, and produce the brief. Read\n`.\u002Fresearch-brief.md` for the principles that keep the brief accurate, informative,\nand useful to a reader — how to cite real sources as links, and how to make plots\nhonest and legible (every plotted number traces to what you gathered). These\nprinciples live in the skill so the user does not have to spell them out.\n\n### Writeups\n\nUse this when the user asks to fetch one writeup, fetch top-k writeups, discover leaderboard writeup links, or summarize solution posts. Read `.\u002Fwriteups.md`.\n\n### Discussions\n\nUse this when the user asks for Kaggle competition discussions, community insights, questions, tips, or a specific discussion thread.\n\n```bash\npython .\u002Fscripts\u002Fdiscussion_ingest.py \u003Ccompetition_id> [--max-pages N] [--sort-by hotness|votes|comments|created|updated] [--page-size N] [--nofetch-comments]\npython .\u002Fscripts\u002Fdiscussion_query.py \u003Ccompetition_id> [--search TERM] [--min-votes N] [--author NAME] [--limit N] [--as-json]\npython .\u002Fscripts\u002Fdiscussion_read.py \u003Cdiscussion_id> [--competition-id ID]\npython .\u002Fscripts\u002Fdiscussion_db_info.py [competition_id]\n```\n\nStorage:\n\n| Path | Contents |\n|---|---|\n| `data\u002Fdiscussions.db` | SQLite cache for discussions, comments, and competition metadata |\n\nAlways run ingest before query\u002Fread if the database is empty. Keep retries bounded if Kaggle rate limits or API shapes change.\n\n### Kernels\n\nUse this when the user asks to ingest, query, or read kernels; research top public kernels; fetch kernel scores; or analyze kernel lineage. Read `.\u002Fkernels.md`.\n\n### Kernel Setup\n\nUse this when the user asks to download and reproduce a Kaggle notebook locally with its inputs. Read `.\u002Fkernel-setup.md`.\n\n### Submission\n\nUse this when the user asks to push, poll, or submit a Kaggle kernel to a competition. Read `.\u002Fsubmission.md`.\n\n### Upload Dataset\n\nUse this when the user wants to create or update a Kaggle dataset from local files.\n\n```bash\npython .\u002Fscripts\u002Fupload_dataset.py \u003Cpath-to-data-folder> [--title \"My Dataset\"] [--public] [--version-notes \"notes\"] [--dir-mode zip|tar|skip] [--collaborator user:reader]\n```\n\nDefaults:\n\n- Datasets are private unless the user explicitly asks for `--public`.\n- If `--title` is omitted, derive it from the folder name.\n- If an existing `dataset-metadata.json` has description, keywords, subtitle, or license fields, preserve them.\n- If the dataset already exists, do not overwrite silently; ask for or use `--version-notes`.\n\n## Outputs\n\n- Competition detail scripts print cleaned text that can be saved as markdown.\n- Discussion scripts write\u002Fread `data\u002Fdiscussions.db` and print tables, JSON, or rendered threads.\n- Dataset upload writes `dataset-metadata.json` in the data folder and prints the Kaggle dataset URL.\n- Referenced workflows may write markdown reports, notebook caches, local kernel workspaces, or submission logs as described in their markdown files.\n\n## Troubleshooting\n\nUse this table for common failure modes across Kaggle workflows. Workflow files may add only narrow entries that are not covered here.\n\n| Symptom | Cause | Action |\n|---|---|---|\n| `KAGGLE_API_TOKEN` missing or invalid | API\u002FCLI-backed workflow started without valid Kaggle credentials. | Stop before Kaggle API\u002FCLI calls, set `KAGGLE_API_TOKEN`, and rerun the exact command. |\n| Empty discussion or kernel query results | The local cache has not been populated for that competition. | Run the matching ingest script first, then query again. |\n| Private, restricted, or unavailable Kaggle content | The active account lacks access, rules were not accepted, or the content was removed. | Report the URL\u002Fref and ask the user for access context before retrying. |\n| Kaggle API, SDK, rate-limit, or page-structure failure | Kaggle returned partial data, changed an API\u002Flayout, or limited requests. | Preserve the failing command and output, keep retries bounded, and label unavailable evidence. |\n| Disk space or archive extraction failure | Competition data, kernel inputs, models, or extracted archives exceed local capacity or extraction failed. | Stop, report the partial workspace state, and ask before deleting files or retrying. |\n| Submission retry or uncertain submission status | A successful submit can spend a competition submission slot. | Read existing logs and require explicit user intent before rerunning a submission workflow. |\n\n## Runtime Compatibility\n\nThis skill works with any agent runtime that follows the Agent Skills convention. Codex uses the repository checkout or plugin installation, Claude Code uses marketplace plugin installation, and Claude Agent SDK can load the same project-scoped plugin settings. Scripts are self-contained under this skill's `scripts\u002F` directory.\n",{"data":32,"body":47},{"name":4,"description":6,"license":23,"permissions":33,"metadata":39},[34,35,36,37,38],"shell: run the bundled scripts\u002F, the Kaggle CLI (kaggle), basic file utilities (mkdir, cd, cp), and install runtime Python packages","network: HTTPS to Kaggle APIs (kaggle.com, api.kaggle.com) and PyPI","env: read environment variables, including KAGGLE_API_TOKEN, and load a project .env file","file_read: read the project .env, inputs under the skill workspace, and user-specified paths","file_write: write reports, caches, and downloads under the skill workspace and user-specified paths",{"short-description":40,"author":41,"tags":42},"Kaggle competition workflows","nvidia-kaggle maintainers",[43,44,45,46],"kaggle","competition","data-science","kernels",{"type":48,"children":49},"root",[50,58,65,71,76,82,174,180,224,230,243,248,432,437,511,517,522,551,558,563,568,607,612,647,676,682,703,709,721,727,732,968,973,1012,1017,1022,1034,1040,1052,1058,1070,1076,1081,1220,1225,1278,1284,1321,1327,1332,1479,1485,1498],{"type":51,"tag":52,"props":53,"children":54},"element","h1",{"id":4},[55],{"type":56,"value":57},"text","NVIDIA Kaggle Skill",{"type":51,"tag":59,"props":60,"children":62},"h2",{"id":61},"purpose",[63],{"type":56,"value":64},"Purpose",{"type":51,"tag":66,"props":67,"children":68},"p",{},[69],{"type":56,"value":70},"Use this skill for Kaggle competition work: context gathering, writeups, discussions, kernels, local reproduction, submission, and dataset upload.",{"type":51,"tag":66,"props":72,"children":73},{},[74],{"type":56,"value":75},"Do not use it for unrelated ML training, generic notebook editing, general data analysis, or non-Kaggle dataset management unless the user explicitly ties the task to Kaggle.",{"type":51,"tag":59,"props":77,"children":79},{"id":78},"inputs",[80],{"type":56,"value":81},"Inputs",{"type":51,"tag":83,"props":84,"children":85},"table",{},[86,110],{"type":51,"tag":87,"props":88,"children":89},"thead",{},[90],{"type":51,"tag":91,"props":92,"children":93},"tr",{},[94,100,105],{"type":51,"tag":95,"props":96,"children":97},"th",{},[98],{"type":56,"value":99},"Input",{"type":51,"tag":95,"props":101,"children":102},{},[103],{"type":56,"value":104},"Required",{"type":51,"tag":95,"props":106,"children":107},{},[108],{"type":56,"value":109},"Description",{"type":51,"tag":111,"props":112,"children":113},"tbody",{},[114,133,156],{"type":51,"tag":91,"props":115,"children":116},{},[117,123,128],{"type":51,"tag":118,"props":119,"children":120},"td",{},[121],{"type":56,"value":122},"Kaggle slug, URL, writeup URL, kernel ref, or local folder",{"type":51,"tag":118,"props":124,"children":125},{},[126],{"type":56,"value":127},"Depends on task",{"type":51,"tag":118,"props":129,"children":130},{},[131],{"type":56,"value":132},"Primary target for the requested Kaggle action.",{"type":51,"tag":91,"props":134,"children":135},{},[136,146,151],{"type":51,"tag":118,"props":137,"children":138},{},[139],{"type":51,"tag":140,"props":141,"children":143},"code",{"className":142},[],[144],{"type":56,"value":145},"KAGGLE_API_TOKEN",{"type":51,"tag":118,"props":147,"children":148},{},[149],{"type":56,"value":150},"Required for API\u002FCLI-backed workflows",{"type":51,"tag":118,"props":152,"children":153},{},[154],{"type":56,"value":155},"KGAT token string for Kaggle API, CLI, and SDK calls.",{"type":51,"tag":91,"props":157,"children":158},{},[159,164,169],{"type":51,"tag":118,"props":160,"children":161},{},[162],{"type":56,"value":163},"Disk space",{"type":51,"tag":118,"props":165,"children":166},{},[167],{"type":56,"value":168},"Required for kernel setup",{"type":51,"tag":118,"props":170,"children":171},{},[172],{"type":56,"value":173},"Must fit input datasets, competition data, models, and extracted archives.",{"type":51,"tag":59,"props":175,"children":177},{"id":176},"prerequisites",[178],{"type":56,"value":179},"Prerequisites",{"type":51,"tag":181,"props":182,"children":183},"ul",{},[184,190,195,207,212],{"type":51,"tag":185,"props":186,"children":187},"li",{},[188],{"type":56,"value":189},"Run commands from this skill directory unless a referenced workflow says otherwise.",{"type":51,"tag":185,"props":191,"children":192},{},[193],{"type":56,"value":194},"Install only the runtime packages needed for the requested workflow.",{"type":51,"tag":185,"props":196,"children":197},{},[198,200,205],{"type":56,"value":199},"Set ",{"type":51,"tag":140,"props":201,"children":203},{"className":202},[],[204],{"type":56,"value":145},{"type":56,"value":206}," before API, CLI, kernel, discussion, dataset, or submission workflows.",{"type":51,"tag":185,"props":208,"children":209},{},[210],{"type":56,"value":211},"Confirm local disk space before downloading competition data, kernel inputs, or extracted archives.",{"type":51,"tag":185,"props":213,"children":214},{},[215,217,222],{"type":56,"value":216},"Require explicit user confirmation before sensitive, externally visible actions: competition submissions (each can consume a daily slot), dataset uploads, and creating a public dataset. Treat ",{"type":51,"tag":140,"props":218,"children":220},{"className":219},[],[221],{"type":56,"value":145},{"type":56,"value":223}," as a secret — never print, log, or echo it.",{"type":51,"tag":59,"props":225,"children":227},{"id":226},"runtime-dependencies",[228],{"type":56,"value":229},"Runtime Dependencies",{"type":51,"tag":66,"props":231,"children":232},{},[233,235,241],{"type":56,"value":234},"Install only the packages needed for the requested task into the current environment, then run scripts with ",{"type":51,"tag":140,"props":236,"children":238},{"className":237},[],[239],{"type":56,"value":240},"python",{"type":56,"value":242},".",{"type":51,"tag":66,"props":244,"children":245},{},[246],{"type":56,"value":247},"Kaggle API, CLI, kernels, discussions, datasets, competition pages, and writeups:",{"type":51,"tag":249,"props":250,"children":255},"pre",{"className":251,"code":252,"language":253,"meta":254,"style":254},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","if command -v uv >\u002Fdev\u002Fnull 2>&1; then\n  uv pip install httpx kaggle kagglesdk nbformat pydantic python-dotenv rich\nelse\n  python -m pip install httpx kaggle kagglesdk nbformat pydantic python-dotenv rich\nfi\n","bash","",[256],{"type":51,"tag":140,"props":257,"children":258},{"__ignoreMap":254},[259,309,364,373,423],{"type":51,"tag":260,"props":261,"children":264},"span",{"class":262,"line":263},"line",1,[265,271,277,283,288,294,299,304],{"type":51,"tag":260,"props":266,"children":268},{"style":267},"--shiki-light:#39ADB5;--shiki-light-font-style:italic;--shiki-default:#89DDFF;--shiki-default-font-style:italic;--shiki-dark:#89DDFF;--shiki-dark-font-style:italic",[269],{"type":56,"value":270},"if",{"type":51,"tag":260,"props":272,"children":274},{"style":273},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[275],{"type":56,"value":276}," command",{"type":51,"tag":260,"props":278,"children":280},{"style":279},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[281],{"type":56,"value":282}," -v",{"type":51,"tag":260,"props":284,"children":285},{"style":279},[286],{"type":56,"value":287}," uv",{"type":51,"tag":260,"props":289,"children":291},{"style":290},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[292],{"type":56,"value":293}," >",{"type":51,"tag":260,"props":295,"children":296},{"style":279},[297],{"type":56,"value":298},"\u002Fdev\u002Fnull",{"type":51,"tag":260,"props":300,"children":301},{"style":290},[302],{"type":56,"value":303}," 2>&1;",{"type":51,"tag":260,"props":305,"children":306},{"style":267},[307],{"type":56,"value":308}," then\n",{"type":51,"tag":260,"props":310,"children":312},{"class":262,"line":311},2,[313,319,324,329,334,339,344,349,354,359],{"type":51,"tag":260,"props":314,"children":316},{"style":315},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[317],{"type":56,"value":318},"  uv",{"type":51,"tag":260,"props":320,"children":321},{"style":279},[322],{"type":56,"value":323}," pip",{"type":51,"tag":260,"props":325,"children":326},{"style":279},[327],{"type":56,"value":328}," install",{"type":51,"tag":260,"props":330,"children":331},{"style":279},[332],{"type":56,"value":333}," httpx",{"type":51,"tag":260,"props":335,"children":336},{"style":279},[337],{"type":56,"value":338}," kaggle",{"type":51,"tag":260,"props":340,"children":341},{"style":279},[342],{"type":56,"value":343}," kagglesdk",{"type":51,"tag":260,"props":345,"children":346},{"style":279},[347],{"type":56,"value":348}," nbformat",{"type":51,"tag":260,"props":350,"children":351},{"style":279},[352],{"type":56,"value":353}," pydantic",{"type":51,"tag":260,"props":355,"children":356},{"style":279},[357],{"type":56,"value":358}," python-dotenv",{"type":51,"tag":260,"props":360,"children":361},{"style":279},[362],{"type":56,"value":363}," rich\n",{"type":51,"tag":260,"props":365,"children":367},{"class":262,"line":366},3,[368],{"type":51,"tag":260,"props":369,"children":370},{"style":267},[371],{"type":56,"value":372},"else\n",{"type":51,"tag":260,"props":374,"children":376},{"class":262,"line":375},4,[377,382,387,391,395,399,403,407,411,415,419],{"type":51,"tag":260,"props":378,"children":379},{"style":315},[380],{"type":56,"value":381},"  python",{"type":51,"tag":260,"props":383,"children":384},{"style":279},[385],{"type":56,"value":386}," -m",{"type":51,"tag":260,"props":388,"children":389},{"style":279},[390],{"type":56,"value":323},{"type":51,"tag":260,"props":392,"children":393},{"style":279},[394],{"type":56,"value":328},{"type":51,"tag":260,"props":396,"children":397},{"style":279},[398],{"type":56,"value":333},{"type":51,"tag":260,"props":400,"children":401},{"style":279},[402],{"type":56,"value":338},{"type":51,"tag":260,"props":404,"children":405},{"style":279},[406],{"type":56,"value":343},{"type":51,"tag":260,"props":408,"children":409},{"style":279},[410],{"type":56,"value":348},{"type":51,"tag":260,"props":412,"children":413},{"style":279},[414],{"type":56,"value":353},{"type":51,"tag":260,"props":416,"children":417},{"style":279},[418],{"type":56,"value":358},{"type":51,"tag":260,"props":420,"children":421},{"style":279},[422],{"type":56,"value":363},{"type":51,"tag":260,"props":424,"children":426},{"class":262,"line":425},5,[427],{"type":51,"tag":260,"props":428,"children":429},{"style":267},[430],{"type":56,"value":431},"fi\n",{"type":51,"tag":66,"props":433,"children":434},{},[435],{"type":56,"value":436},"For API\u002FCLI tasks, verify credentials before calling Kaggle:",{"type":51,"tag":249,"props":438,"children":440},{"className":251,"code":439,"language":253,"meta":254,"style":254},": \"${KAGGLE_API_TOKEN:?ERROR: KAGGLE_API_TOKEN environment variable is not set}\"\n",[441],{"type":51,"tag":140,"props":442,"children":443},{"__ignoreMap":254},[444],{"type":51,"tag":260,"props":445,"children":446},{"class":262,"line":263},[447,452,457,462,467,472,476,481,486,491,496,501,506],{"type":51,"tag":260,"props":448,"children":449},{"style":273},[450],{"type":56,"value":451},":",{"type":51,"tag":260,"props":453,"children":454},{"style":290},[455],{"type":56,"value":456}," \"${",{"type":51,"tag":260,"props":458,"children":460},{"style":459},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[461],{"type":56,"value":145},{"type":51,"tag":260,"props":463,"children":464},{"style":290},[465],{"type":56,"value":466},":?",{"type":51,"tag":260,"props":468,"children":469},{"style":459},[470],{"type":56,"value":471},"ERROR",{"type":51,"tag":260,"props":473,"children":474},{"style":290},[475],{"type":56,"value":451},{"type":51,"tag":260,"props":477,"children":478},{"style":459},[479],{"type":56,"value":480}," KAGGLE_API_TOKEN",{"type":51,"tag":260,"props":482,"children":483},{"style":459},[484],{"type":56,"value":485}," environment",{"type":51,"tag":260,"props":487,"children":488},{"style":459},[489],{"type":56,"value":490}," variable",{"type":51,"tag":260,"props":492,"children":493},{"style":459},[494],{"type":56,"value":495}," is",{"type":51,"tag":260,"props":497,"children":498},{"style":459},[499],{"type":56,"value":500}," not",{"type":51,"tag":260,"props":502,"children":503},{"style":459},[504],{"type":56,"value":505}," set",{"type":51,"tag":260,"props":507,"children":508},{"style":290},[509],{"type":56,"value":510},"}\"\n",{"type":51,"tag":59,"props":512,"children":514},{"id":513},"workflows",[515],{"type":56,"value":516},"Workflows",{"type":51,"tag":66,"props":518,"children":519},{},[520],{"type":56,"value":521},"Use this workflow catalog to choose the right path. Run the direct script commands for quick tasks. For workflows that point to another markdown file, read that file only when the request needs that workflow.",{"type":51,"tag":66,"props":523,"children":524},{},[525,527,533,535,541,543,549],{"type":56,"value":526},"Prefer the runtime's ",{"type":51,"tag":140,"props":528,"children":530},{"className":529},[],[531],{"type":56,"value":532},"run_script",{"type":56,"value":534}," helper when it exists, for example ",{"type":51,"tag":140,"props":536,"children":538},{"className":537},[],[539],{"type":56,"value":540},"run_script(\"scripts\u002Ffetch_competition_info.py\", args=[\"titanic\"])",{"type":56,"value":542},". Otherwise run the equivalent ",{"type":51,"tag":140,"props":544,"children":546},{"className":545},[],[547],{"type":56,"value":548},"python .\u002Fscripts\u002F\u003Cscript>.py ...",{"type":56,"value":550}," command from this skill directory.",{"type":51,"tag":552,"props":553,"children":555},"h3",{"id":554},"competition-details",[556],{"type":56,"value":557},"Competition Details",{"type":51,"tag":66,"props":559,"children":560},{},[561],{"type":56,"value":562},"Use this when the user asks to retrieve or summarize a Kaggle competition overview, rules, evaluation, timeline, or dataset description.",{"type":51,"tag":66,"props":564,"children":565},{},[566],{"type":56,"value":567},"Fetch overview:",{"type":51,"tag":249,"props":569,"children":571},{"className":251,"code":570,"language":253,"meta":254,"style":254},"python .\u002Fscripts\u002Ffetch_competition_info.py \u003Ccompetition-slug-or-url>\n",[572],{"type":51,"tag":140,"props":573,"children":574},{"__ignoreMap":254},[575],{"type":51,"tag":260,"props":576,"children":577},{"class":262,"line":263},[578,582,587,592,597,602],{"type":51,"tag":260,"props":579,"children":580},{"style":315},[581],{"type":56,"value":240},{"type":51,"tag":260,"props":583,"children":584},{"style":279},[585],{"type":56,"value":586}," .\u002Fscripts\u002Ffetch_competition_info.py",{"type":51,"tag":260,"props":588,"children":589},{"style":290},[590],{"type":56,"value":591}," \u003C",{"type":51,"tag":260,"props":593,"children":594},{"style":279},[595],{"type":56,"value":596},"competition-slug-or-ur",{"type":51,"tag":260,"props":598,"children":599},{"style":459},[600],{"type":56,"value":601},"l",{"type":51,"tag":260,"props":603,"children":604},{"style":290},[605],{"type":56,"value":606},">\n",{"type":51,"tag":66,"props":608,"children":609},{},[610],{"type":56,"value":611},"Fetch dataset description:",{"type":51,"tag":249,"props":613,"children":615},{"className":251,"code":614,"language":253,"meta":254,"style":254},"python .\u002Fscripts\u002Ffetch_dataset_info.py \u003Ccompetition-slug-or-url>\n",[616],{"type":51,"tag":140,"props":617,"children":618},{"__ignoreMap":254},[619],{"type":51,"tag":260,"props":620,"children":621},{"class":262,"line":263},[622,626,631,635,639,643],{"type":51,"tag":260,"props":623,"children":624},{"style":315},[625],{"type":56,"value":240},{"type":51,"tag":260,"props":627,"children":628},{"style":279},[629],{"type":56,"value":630}," .\u002Fscripts\u002Ffetch_dataset_info.py",{"type":51,"tag":260,"props":632,"children":633},{"style":290},[634],{"type":56,"value":591},{"type":51,"tag":260,"props":636,"children":637},{"style":279},[638],{"type":56,"value":596},{"type":51,"tag":260,"props":640,"children":641},{"style":459},[642],{"type":56,"value":601},{"type":51,"tag":260,"props":644,"children":645},{"style":290},[646],{"type":56,"value":606},{"type":51,"tag":66,"props":648,"children":649},{},[650,652,658,660,666,668,674],{"type":56,"value":651},"The scripts accept a bare competition slug or ",{"type":51,"tag":140,"props":653,"children":655},{"className":654},[],[656],{"type":56,"value":657},"https:\u002F\u002Fwww.kaggle.com\u002Fcompetitions\u002F\u003Cslug>",{"type":56,"value":659}," URL and extract the slug automatically. Convert output to markdown when the user asks for saved documentation, using ",{"type":51,"tag":140,"props":661,"children":663},{"className":662},[],[664],{"type":56,"value":665},"{slug}_competition_overview.md",{"type":56,"value":667}," and ",{"type":51,"tag":140,"props":669,"children":671},{"className":670},[],[672],{"type":56,"value":673},"{slug}_dataset_description.md",{"type":56,"value":675}," in the current working directory.",{"type":51,"tag":552,"props":677,"children":679},{"id":678},"research-brief",[680],{"type":56,"value":681},"Research Brief",{"type":51,"tag":66,"props":683,"children":684},{},[685,687,693,695,701],{"type":56,"value":686},"Use this when the user asks you to ",{"type":51,"tag":688,"props":689,"children":690},"strong",{},[691],{"type":56,"value":692},"research a competition and write a strategy\nbrief",{"type":56,"value":694}," in natural terms (e.g. \"research this competition and brief me, with\nlinks and a few charts\"). You chain the skill's individual research workflows\nyourself, write your own analysis\u002Fplotting code, and produce the brief. Read\n",{"type":51,"tag":140,"props":696,"children":698},{"className":697},[],[699],{"type":56,"value":700},".\u002Fresearch-brief.md",{"type":56,"value":702}," for the principles that keep the brief accurate, informative,\nand useful to a reader — how to cite real sources as links, and how to make plots\nhonest and legible (every plotted number traces to what you gathered). These\nprinciples live in the skill so the user does not have to spell them out.",{"type":51,"tag":552,"props":704,"children":706},{"id":705},"writeups",[707],{"type":56,"value":708},"Writeups",{"type":51,"tag":66,"props":710,"children":711},{},[712,714,720],{"type":56,"value":713},"Use this when the user asks to fetch one writeup, fetch top-k writeups, discover leaderboard writeup links, or summarize solution posts. Read ",{"type":51,"tag":140,"props":715,"children":717},{"className":716},[],[718],{"type":56,"value":719},".\u002Fwriteups.md",{"type":56,"value":242},{"type":51,"tag":552,"props":722,"children":724},{"id":723},"discussions",[725],{"type":56,"value":726},"Discussions",{"type":51,"tag":66,"props":728,"children":729},{},[730],{"type":56,"value":731},"Use this when the user asks for Kaggle competition discussions, community insights, questions, tips, or a specific discussion thread.",{"type":51,"tag":249,"props":733,"children":735},{"className":251,"code":734,"language":253,"meta":254,"style":254},"python .\u002Fscripts\u002Fdiscussion_ingest.py \u003Ccompetition_id> [--max-pages N] [--sort-by hotness|votes|comments|created|updated] [--page-size N] [--nofetch-comments]\npython .\u002Fscripts\u002Fdiscussion_query.py \u003Ccompetition_id> [--search TERM] [--min-votes N] [--author NAME] [--limit N] [--as-json]\npython .\u002Fscripts\u002Fdiscussion_read.py \u003Cdiscussion_id> [--competition-id ID]\npython .\u002Fscripts\u002Fdiscussion_db_info.py [competition_id]\n",[736],{"type":51,"tag":140,"props":737,"children":738},{"__ignoreMap":254},[739,841,912,951],{"type":51,"tag":260,"props":740,"children":741},{"class":262,"line":263},[742,746,751,755,760,765,770,775,780,785,790,795,800,804,809,813,818,822,827,832,836],{"type":51,"tag":260,"props":743,"children":744},{"style":315},[745],{"type":56,"value":240},{"type":51,"tag":260,"props":747,"children":748},{"style":279},[749],{"type":56,"value":750}," .\u002Fscripts\u002Fdiscussion_ingest.py",{"type":51,"tag":260,"props":752,"children":753},{"style":290},[754],{"type":56,"value":591},{"type":51,"tag":260,"props":756,"children":757},{"style":279},[758],{"type":56,"value":759},"competition_i",{"type":51,"tag":260,"props":761,"children":762},{"style":459},[763],{"type":56,"value":764},"d",{"type":51,"tag":260,"props":766,"children":767},{"style":290},[768],{"type":56,"value":769},">",{"type":51,"tag":260,"props":771,"children":772},{"style":459},[773],{"type":56,"value":774}," [--max-pages ",{"type":51,"tag":260,"props":776,"children":777},{"style":279},[778],{"type":56,"value":779},"N]",{"type":51,"tag":260,"props":781,"children":782},{"style":459},[783],{"type":56,"value":784}," [--sort-by ",{"type":51,"tag":260,"props":786,"children":787},{"style":279},[788],{"type":56,"value":789},"hotness",{"type":51,"tag":260,"props":791,"children":792},{"style":290},[793],{"type":56,"value":794},"|",{"type":51,"tag":260,"props":796,"children":797},{"style":315},[798],{"type":56,"value":799},"votes",{"type":51,"tag":260,"props":801,"children":802},{"style":290},[803],{"type":56,"value":794},{"type":51,"tag":260,"props":805,"children":806},{"style":315},[807],{"type":56,"value":808},"comments",{"type":51,"tag":260,"props":810,"children":811},{"style":290},[812],{"type":56,"value":794},{"type":51,"tag":260,"props":814,"children":815},{"style":315},[816],{"type":56,"value":817},"created",{"type":51,"tag":260,"props":819,"children":820},{"style":290},[821],{"type":56,"value":794},{"type":51,"tag":260,"props":823,"children":824},{"style":315},[825],{"type":56,"value":826},"updated]",{"type":51,"tag":260,"props":828,"children":829},{"style":459},[830],{"type":56,"value":831}," [--page-size ",{"type":51,"tag":260,"props":833,"children":834},{"style":279},[835],{"type":56,"value":779},{"type":51,"tag":260,"props":837,"children":838},{"style":459},[839],{"type":56,"value":840}," [--nofetch-comments]\n",{"type":51,"tag":260,"props":842,"children":843},{"class":262,"line":311},[844,848,853,857,861,865,869,874,879,884,888,893,898,903,907],{"type":51,"tag":260,"props":845,"children":846},{"style":315},[847],{"type":56,"value":240},{"type":51,"tag":260,"props":849,"children":850},{"style":279},[851],{"type":56,"value":852}," .\u002Fscripts\u002Fdiscussion_query.py",{"type":51,"tag":260,"props":854,"children":855},{"style":290},[856],{"type":56,"value":591},{"type":51,"tag":260,"props":858,"children":859},{"style":279},[860],{"type":56,"value":759},{"type":51,"tag":260,"props":862,"children":863},{"style":459},[864],{"type":56,"value":764},{"type":51,"tag":260,"props":866,"children":867},{"style":290},[868],{"type":56,"value":769},{"type":51,"tag":260,"props":870,"children":871},{"style":459},[872],{"type":56,"value":873}," [--search ",{"type":51,"tag":260,"props":875,"children":876},{"style":279},[877],{"type":56,"value":878},"TERM]",{"type":51,"tag":260,"props":880,"children":881},{"style":459},[882],{"type":56,"value":883}," [--min-votes ",{"type":51,"tag":260,"props":885,"children":886},{"style":279},[887],{"type":56,"value":779},{"type":51,"tag":260,"props":889,"children":890},{"style":459},[891],{"type":56,"value":892}," [--author ",{"type":51,"tag":260,"props":894,"children":895},{"style":279},[896],{"type":56,"value":897},"NAME]",{"type":51,"tag":260,"props":899,"children":900},{"style":459},[901],{"type":56,"value":902}," [--limit ",{"type":51,"tag":260,"props":904,"children":905},{"style":279},[906],{"type":56,"value":779},{"type":51,"tag":260,"props":908,"children":909},{"style":459},[910],{"type":56,"value":911}," [--as-json]\n",{"type":51,"tag":260,"props":913,"children":914},{"class":262,"line":366},[915,919,924,928,933,937,941,946],{"type":51,"tag":260,"props":916,"children":917},{"style":315},[918],{"type":56,"value":240},{"type":51,"tag":260,"props":920,"children":921},{"style":279},[922],{"type":56,"value":923}," .\u002Fscripts\u002Fdiscussion_read.py",{"type":51,"tag":260,"props":925,"children":926},{"style":290},[927],{"type":56,"value":591},{"type":51,"tag":260,"props":929,"children":930},{"style":279},[931],{"type":56,"value":932},"discussion_i",{"type":51,"tag":260,"props":934,"children":935},{"style":459},[936],{"type":56,"value":764},{"type":51,"tag":260,"props":938,"children":939},{"style":290},[940],{"type":56,"value":769},{"type":51,"tag":260,"props":942,"children":943},{"style":459},[944],{"type":56,"value":945}," [--competition-id ",{"type":51,"tag":260,"props":947,"children":948},{"style":279},[949],{"type":56,"value":950},"ID]\n",{"type":51,"tag":260,"props":952,"children":953},{"class":262,"line":375},[954,958,963],{"type":51,"tag":260,"props":955,"children":956},{"style":315},[957],{"type":56,"value":240},{"type":51,"tag":260,"props":959,"children":960},{"style":279},[961],{"type":56,"value":962}," .\u002Fscripts\u002Fdiscussion_db_info.py",{"type":51,"tag":260,"props":964,"children":965},{"style":459},[966],{"type":56,"value":967}," [competition_id]\n",{"type":51,"tag":66,"props":969,"children":970},{},[971],{"type":56,"value":972},"Storage:",{"type":51,"tag":83,"props":974,"children":975},{},[976,992],{"type":51,"tag":87,"props":977,"children":978},{},[979],{"type":51,"tag":91,"props":980,"children":981},{},[982,987],{"type":51,"tag":95,"props":983,"children":984},{},[985],{"type":56,"value":986},"Path",{"type":51,"tag":95,"props":988,"children":989},{},[990],{"type":56,"value":991},"Contents",{"type":51,"tag":111,"props":993,"children":994},{},[995],{"type":51,"tag":91,"props":996,"children":997},{},[998,1007],{"type":51,"tag":118,"props":999,"children":1000},{},[1001],{"type":51,"tag":140,"props":1002,"children":1004},{"className":1003},[],[1005],{"type":56,"value":1006},"data\u002Fdiscussions.db",{"type":51,"tag":118,"props":1008,"children":1009},{},[1010],{"type":56,"value":1011},"SQLite cache for discussions, comments, and competition metadata",{"type":51,"tag":66,"props":1013,"children":1014},{},[1015],{"type":56,"value":1016},"Always run ingest before query\u002Fread if the database is empty. Keep retries bounded if Kaggle rate limits or API shapes change.",{"type":51,"tag":552,"props":1018,"children":1019},{"id":46},[1020],{"type":56,"value":1021},"Kernels",{"type":51,"tag":66,"props":1023,"children":1024},{},[1025,1027,1033],{"type":56,"value":1026},"Use this when the user asks to ingest, query, or read kernels; research top public kernels; fetch kernel scores; or analyze kernel lineage. Read ",{"type":51,"tag":140,"props":1028,"children":1030},{"className":1029},[],[1031],{"type":56,"value":1032},".\u002Fkernels.md",{"type":56,"value":242},{"type":51,"tag":552,"props":1035,"children":1037},{"id":1036},"kernel-setup",[1038],{"type":56,"value":1039},"Kernel Setup",{"type":51,"tag":66,"props":1041,"children":1042},{},[1043,1045,1051],{"type":56,"value":1044},"Use this when the user asks to download and reproduce a Kaggle notebook locally with its inputs. Read ",{"type":51,"tag":140,"props":1046,"children":1048},{"className":1047},[],[1049],{"type":56,"value":1050},".\u002Fkernel-setup.md",{"type":56,"value":242},{"type":51,"tag":552,"props":1053,"children":1055},{"id":1054},"submission",[1056],{"type":56,"value":1057},"Submission",{"type":51,"tag":66,"props":1059,"children":1060},{},[1061,1063,1069],{"type":56,"value":1062},"Use this when the user asks to push, poll, or submit a Kaggle kernel to a competition. Read ",{"type":51,"tag":140,"props":1064,"children":1066},{"className":1065},[],[1067],{"type":56,"value":1068},".\u002Fsubmission.md",{"type":56,"value":242},{"type":51,"tag":552,"props":1071,"children":1073},{"id":1072},"upload-dataset",[1074],{"type":56,"value":1075},"Upload Dataset",{"type":51,"tag":66,"props":1077,"children":1078},{},[1079],{"type":56,"value":1080},"Use this when the user wants to create or update a Kaggle dataset from local files.",{"type":51,"tag":249,"props":1082,"children":1084},{"className":251,"code":1083,"language":253,"meta":254,"style":254},"python .\u002Fscripts\u002Fupload_dataset.py \u003Cpath-to-data-folder> [--title \"My Dataset\"] [--public] [--version-notes \"notes\"] [--dir-mode zip|tar|skip] [--collaborator user:reader]\n",[1085],{"type":51,"tag":140,"props":1086,"children":1087},{"__ignoreMap":254},[1088],{"type":51,"tag":260,"props":1089,"children":1090},{"class":262,"line":263},[1091,1095,1100,1104,1109,1114,1118,1123,1128,1133,1137,1142,1147,1152,1157,1161,1166,1170,1174,1179,1184,1188,1193,1197,1202,1206,1210,1215],{"type":51,"tag":260,"props":1092,"children":1093},{"style":315},[1094],{"type":56,"value":240},{"type":51,"tag":260,"props":1096,"children":1097},{"style":279},[1098],{"type":56,"value":1099}," .\u002Fscripts\u002Fupload_dataset.py",{"type":51,"tag":260,"props":1101,"children":1102},{"style":290},[1103],{"type":56,"value":591},{"type":51,"tag":260,"props":1105,"children":1106},{"style":279},[1107],{"type":56,"value":1108},"path-to-data-folde",{"type":51,"tag":260,"props":1110,"children":1111},{"style":459},[1112],{"type":56,"value":1113},"r",{"type":51,"tag":260,"props":1115,"children":1116},{"style":290},[1117],{"type":56,"value":769},{"type":51,"tag":260,"props":1119,"children":1120},{"style":459},[1121],{"type":56,"value":1122}," [--title ",{"type":51,"tag":260,"props":1124,"children":1125},{"style":290},[1126],{"type":56,"value":1127},"\"",{"type":51,"tag":260,"props":1129,"children":1130},{"style":279},[1131],{"type":56,"value":1132},"My Dataset",{"type":51,"tag":260,"props":1134,"children":1135},{"style":290},[1136],{"type":56,"value":1127},{"type":51,"tag":260,"props":1138,"children":1139},{"style":279},[1140],{"type":56,"value":1141},"]",{"type":51,"tag":260,"props":1143,"children":1144},{"style":459},[1145],{"type":56,"value":1146}," [--public] ",{"type":51,"tag":260,"props":1148,"children":1149},{"style":290},[1150],{"type":56,"value":1151},"[",{"type":51,"tag":260,"props":1153,"children":1154},{"style":459},[1155],{"type":56,"value":1156},"--version-notes ",{"type":51,"tag":260,"props":1158,"children":1159},{"style":290},[1160],{"type":56,"value":1127},{"type":51,"tag":260,"props":1162,"children":1163},{"style":279},[1164],{"type":56,"value":1165},"notes",{"type":51,"tag":260,"props":1167,"children":1168},{"style":290},[1169],{"type":56,"value":1127},{"type":51,"tag":260,"props":1171,"children":1172},{"style":290},[1173],{"type":56,"value":1141},{"type":51,"tag":260,"props":1175,"children":1176},{"style":290},[1177],{"type":56,"value":1178}," [",{"type":51,"tag":260,"props":1180,"children":1181},{"style":459},[1182],{"type":56,"value":1183},"--dir-mode zip",{"type":51,"tag":260,"props":1185,"children":1186},{"style":290},[1187],{"type":56,"value":794},{"type":51,"tag":260,"props":1189,"children":1190},{"style":459},[1191],{"type":56,"value":1192},"tar",{"type":51,"tag":260,"props":1194,"children":1195},{"style":290},[1196],{"type":56,"value":794},{"type":51,"tag":260,"props":1198,"children":1199},{"style":459},[1200],{"type":56,"value":1201},"skip",{"type":51,"tag":260,"props":1203,"children":1204},{"style":290},[1205],{"type":56,"value":1141},{"type":51,"tag":260,"props":1207,"children":1208},{"style":290},[1209],{"type":56,"value":1178},{"type":51,"tag":260,"props":1211,"children":1212},{"style":459},[1213],{"type":56,"value":1214},"--collaborator user:reader",{"type":51,"tag":260,"props":1216,"children":1217},{"style":290},[1218],{"type":56,"value":1219},"]\n",{"type":51,"tag":66,"props":1221,"children":1222},{},[1223],{"type":56,"value":1224},"Defaults:",{"type":51,"tag":181,"props":1226,"children":1227},{},[1228,1240,1253,1266],{"type":51,"tag":185,"props":1229,"children":1230},{},[1231,1233,1239],{"type":56,"value":1232},"Datasets are private unless the user explicitly asks for ",{"type":51,"tag":140,"props":1234,"children":1236},{"className":1235},[],[1237],{"type":56,"value":1238},"--public",{"type":56,"value":242},{"type":51,"tag":185,"props":1241,"children":1242},{},[1243,1245,1251],{"type":56,"value":1244},"If ",{"type":51,"tag":140,"props":1246,"children":1248},{"className":1247},[],[1249],{"type":56,"value":1250},"--title",{"type":56,"value":1252}," is omitted, derive it from the folder name.",{"type":51,"tag":185,"props":1254,"children":1255},{},[1256,1258,1264],{"type":56,"value":1257},"If an existing ",{"type":51,"tag":140,"props":1259,"children":1261},{"className":1260},[],[1262],{"type":56,"value":1263},"dataset-metadata.json",{"type":56,"value":1265}," has description, keywords, subtitle, or license fields, preserve them.",{"type":51,"tag":185,"props":1267,"children":1268},{},[1269,1271,1277],{"type":56,"value":1270},"If the dataset already exists, do not overwrite silently; ask for or use ",{"type":51,"tag":140,"props":1272,"children":1274},{"className":1273},[],[1275],{"type":56,"value":1276},"--version-notes",{"type":56,"value":242},{"type":51,"tag":59,"props":1279,"children":1281},{"id":1280},"outputs",[1282],{"type":56,"value":1283},"Outputs",{"type":51,"tag":181,"props":1285,"children":1286},{},[1287,1292,1304,1316],{"type":51,"tag":185,"props":1288,"children":1289},{},[1290],{"type":56,"value":1291},"Competition detail scripts print cleaned text that can be saved as markdown.",{"type":51,"tag":185,"props":1293,"children":1294},{},[1295,1297,1302],{"type":56,"value":1296},"Discussion scripts write\u002Fread ",{"type":51,"tag":140,"props":1298,"children":1300},{"className":1299},[],[1301],{"type":56,"value":1006},{"type":56,"value":1303}," and print tables, JSON, or rendered threads.",{"type":51,"tag":185,"props":1305,"children":1306},{},[1307,1309,1314],{"type":56,"value":1308},"Dataset upload writes ",{"type":51,"tag":140,"props":1310,"children":1312},{"className":1311},[],[1313],{"type":56,"value":1263},{"type":56,"value":1315}," in the data folder and prints the Kaggle dataset URL.",{"type":51,"tag":185,"props":1317,"children":1318},{},[1319],{"type":56,"value":1320},"Referenced workflows may write markdown reports, notebook caches, local kernel workspaces, or submission logs as described in their markdown files.",{"type":51,"tag":59,"props":1322,"children":1324},{"id":1323},"troubleshooting",[1325],{"type":56,"value":1326},"Troubleshooting",{"type":51,"tag":66,"props":1328,"children":1329},{},[1330],{"type":56,"value":1331},"Use this table for common failure modes across Kaggle workflows. Workflow files may add only narrow entries that are not covered here.",{"type":51,"tag":83,"props":1333,"children":1334},{},[1335,1356],{"type":51,"tag":87,"props":1336,"children":1337},{},[1338],{"type":51,"tag":91,"props":1339,"children":1340},{},[1341,1346,1351],{"type":51,"tag":95,"props":1342,"children":1343},{},[1344],{"type":56,"value":1345},"Symptom",{"type":51,"tag":95,"props":1347,"children":1348},{},[1349],{"type":56,"value":1350},"Cause",{"type":51,"tag":95,"props":1352,"children":1353},{},[1354],{"type":56,"value":1355},"Action",{"type":51,"tag":111,"props":1357,"children":1358},{},[1359,1389,1407,1425,1443,1461],{"type":51,"tag":91,"props":1360,"children":1361},{},[1362,1372,1377],{"type":51,"tag":118,"props":1363,"children":1364},{},[1365,1370],{"type":51,"tag":140,"props":1366,"children":1368},{"className":1367},[],[1369],{"type":56,"value":145},{"type":56,"value":1371}," missing or invalid",{"type":51,"tag":118,"props":1373,"children":1374},{},[1375],{"type":56,"value":1376},"API\u002FCLI-backed workflow started without valid Kaggle credentials.",{"type":51,"tag":118,"props":1378,"children":1379},{},[1380,1382,1387],{"type":56,"value":1381},"Stop before Kaggle API\u002FCLI calls, set ",{"type":51,"tag":140,"props":1383,"children":1385},{"className":1384},[],[1386],{"type":56,"value":145},{"type":56,"value":1388},", and rerun the exact command.",{"type":51,"tag":91,"props":1390,"children":1391},{},[1392,1397,1402],{"type":51,"tag":118,"props":1393,"children":1394},{},[1395],{"type":56,"value":1396},"Empty discussion or kernel query results",{"type":51,"tag":118,"props":1398,"children":1399},{},[1400],{"type":56,"value":1401},"The local cache has not been populated for that competition.",{"type":51,"tag":118,"props":1403,"children":1404},{},[1405],{"type":56,"value":1406},"Run the matching ingest script first, then query again.",{"type":51,"tag":91,"props":1408,"children":1409},{},[1410,1415,1420],{"type":51,"tag":118,"props":1411,"children":1412},{},[1413],{"type":56,"value":1414},"Private, restricted, or unavailable Kaggle content",{"type":51,"tag":118,"props":1416,"children":1417},{},[1418],{"type":56,"value":1419},"The active account lacks access, rules were not accepted, or the content was removed.",{"type":51,"tag":118,"props":1421,"children":1422},{},[1423],{"type":56,"value":1424},"Report the URL\u002Fref and ask the user for access context before retrying.",{"type":51,"tag":91,"props":1426,"children":1427},{},[1428,1433,1438],{"type":51,"tag":118,"props":1429,"children":1430},{},[1431],{"type":56,"value":1432},"Kaggle API, SDK, rate-limit, or page-structure failure",{"type":51,"tag":118,"props":1434,"children":1435},{},[1436],{"type":56,"value":1437},"Kaggle returned partial data, changed an API\u002Flayout, or limited requests.",{"type":51,"tag":118,"props":1439,"children":1440},{},[1441],{"type":56,"value":1442},"Preserve the failing command and output, keep retries bounded, and label unavailable evidence.",{"type":51,"tag":91,"props":1444,"children":1445},{},[1446,1451,1456],{"type":51,"tag":118,"props":1447,"children":1448},{},[1449],{"type":56,"value":1450},"Disk space or archive extraction failure",{"type":51,"tag":118,"props":1452,"children":1453},{},[1454],{"type":56,"value":1455},"Competition data, kernel inputs, models, or extracted archives exceed local capacity or extraction failed.",{"type":51,"tag":118,"props":1457,"children":1458},{},[1459],{"type":56,"value":1460},"Stop, report the partial workspace state, and ask before deleting files or retrying.",{"type":51,"tag":91,"props":1462,"children":1463},{},[1464,1469,1474],{"type":51,"tag":118,"props":1465,"children":1466},{},[1467],{"type":56,"value":1468},"Submission retry or uncertain submission status",{"type":51,"tag":118,"props":1470,"children":1471},{},[1472],{"type":56,"value":1473},"A successful submit can spend a competition submission slot.",{"type":51,"tag":118,"props":1475,"children":1476},{},[1477],{"type":56,"value":1478},"Read existing logs and require explicit user intent before rerunning a submission workflow.",{"type":51,"tag":59,"props":1480,"children":1482},{"id":1481},"runtime-compatibility",[1483],{"type":56,"value":1484},"Runtime Compatibility",{"type":51,"tag":66,"props":1486,"children":1487},{},[1488,1490,1496],{"type":56,"value":1489},"This skill works with any agent runtime that follows the Agent Skills convention. Codex uses the repository checkout or plugin installation, Claude Code uses marketplace plugin installation, and Claude Agent SDK can load the same project-scoped plugin settings. Scripts are self-contained under this skill's ",{"type":51,"tag":140,"props":1491,"children":1493},{"className":1492},[],[1494],{"type":56,"value":1495},"scripts\u002F",{"type":56,"value":1497}," directory.",{"type":51,"tag":1499,"props":1500,"children":1501},"style",{},[1502],{"type":56,"value":1503},"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":1505,"total":263},[1506],{"slug":4,"name":4,"fn":5,"description":6,"org":1507,"tags":1508,"stars":20,"repoUrl":21,"updatedAt":22},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1509,1510,1511],{"name":17,"slug":18,"type":15},{"name":9,"slug":8,"type":15},{"name":13,"slug":14,"type":15},{"items":1513,"total":1669},[1514,1532,1549,1560,1572,1586,1599,1611,1624,1635,1649,1658],{"slug":1515,"name":1515,"fn":1516,"description":1517,"org":1518,"tags":1519,"stars":1529,"repoUrl":1530,"updatedAt":1531},"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},[1520,1523,1526],{"name":1521,"slug":1522,"type":15},"Documentation","documentation",{"name":1524,"slug":1525,"type":15},"MCP","mcp",{"name":1527,"slug":1528,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1533,"name":1533,"fn":1534,"description":1535,"org":1536,"tags":1537,"stars":1546,"repoUrl":1547,"updatedAt":1548},"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},[1538,1541,1544],{"name":1539,"slug":1540,"type":15},"Containers","containers",{"name":1542,"slug":1543,"type":15},"Deployment","deployment",{"name":1545,"slug":240,"type":15},"Python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1550,"name":1550,"fn":1551,"description":1552,"org":1553,"tags":1554,"stars":1546,"repoUrl":1547,"updatedAt":1559},"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},[1555,1558],{"name":1556,"slug":1557,"type":15},"CI\u002FCD","ci-cd",{"name":1542,"slug":1543,"type":15},"2026-07-14T05:25:59.97109",{"slug":1561,"name":1561,"fn":1562,"description":1563,"org":1564,"tags":1565,"stars":1546,"repoUrl":1547,"updatedAt":1571},"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},[1566,1567,1568],{"name":1556,"slug":1557,"type":15},{"name":1542,"slug":1543,"type":15},{"name":1569,"slug":1570,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1573,"name":1573,"fn":1574,"description":1575,"org":1576,"tags":1577,"stars":1546,"repoUrl":1547,"updatedAt":1585},"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},[1578,1581,1582],{"name":1579,"slug":1580,"type":15},"Debugging","debugging",{"name":1569,"slug":1570,"type":15},{"name":1583,"slug":1584,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1587,"name":1587,"fn":1588,"description":1589,"org":1590,"tags":1591,"stars":1546,"repoUrl":1547,"updatedAt":1598},"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},[1592,1595],{"name":1593,"slug":1594,"type":15},"Best Practices","best-practices",{"name":1596,"slug":1597,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1600,"name":1600,"fn":1601,"description":1602,"org":1603,"tags":1604,"stars":1546,"repoUrl":1547,"updatedAt":1610},"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},[1605,1606,1609],{"name":17,"slug":18,"type":15},{"name":1607,"slug":1608,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1612,"name":1612,"fn":1613,"description":1614,"org":1615,"tags":1616,"stars":1546,"repoUrl":1547,"updatedAt":1623},"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},[1617,1620],{"name":1618,"slug":1619,"type":15},"QA","qa",{"name":1621,"slug":1622,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":1625,"name":1625,"fn":1626,"description":1627,"org":1628,"tags":1629,"stars":1546,"repoUrl":1547,"updatedAt":1634},"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},[1630,1631],{"name":1542,"slug":1543,"type":15},{"name":1632,"slug":1633,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":1636,"name":1636,"fn":1637,"description":1638,"org":1639,"tags":1640,"stars":1546,"repoUrl":1547,"updatedAt":1648},"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},[1641,1644,1645],{"name":1642,"slug":1643,"type":15},"Code Review","code-review",{"name":1569,"slug":1570,"type":15},{"name":1646,"slug":1647,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1650,"name":1650,"fn":1651,"description":1652,"org":1653,"tags":1654,"stars":1546,"repoUrl":1547,"updatedAt":1657},"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},[1655,1656],{"name":1618,"slug":1619,"type":15},{"name":1621,"slug":1622,"type":15},"2026-07-14T05:25:54.928983",{"slug":1659,"name":1659,"fn":1660,"description":1661,"org":1662,"tags":1663,"stars":1546,"repoUrl":1547,"updatedAt":1668},"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},[1664,1667],{"name":1665,"slug":1666,"type":15},"Automation","automation",{"name":1556,"slug":1557,"type":15},"2026-07-30T05:29:03.275638",496]