[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-skill-card-generator":3,"mdc--rtw5yk-key":34,"related-repo-nvidia-skill-card-generator":856,"related-org-nvidia-skill-card-generator":962},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":29,"sourceUrl":32,"mdContent":33},"skill-card-generator","generate governance skill cards","Use only to generate or update a governance skill card for a specified existing agent skill directory. Do not use for explaining, listing, comparing, or discussing skill capabilities.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,20],{"name":13,"slug":14,"type":15},"Governance","governance","tag",{"name":17,"slug":18,"type":15},"Documentation","documentation",{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},"Technical Writing","technical-writing",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-14T05:28:05.566753","CC-BY-4.0 AND Apache-2.0",281,[],{"repoUrl":24,"stars":23,"forks":27,"topics":30,"description":31},[],"AI agent skills published by NVIDIA","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fskill-card-generator","---\nname: \"skill-card-generator\"\ndescription: \"Use only to generate or update a governance skill card for a specified existing agent skill directory. Do not use for explaining, listing, comparing, or discussing skill capabilities.\"\nlicense: CC-BY-4.0 AND Apache-2.0\ncompatibility: \"Any agent that can run Python scripts and write files\"\nmetadata:\n  author: \"Trustworthy AI Projects \u003Ctrustworthyaiprojects@nvidia.com>\"\n  tags:\n    - skill-card\n    - governance\n    - documentation\n    - trustworthy-ai\n  domain: documentation\npermissions:\n  file_read:\n    - \"target_skill_directory\"\n    - \"references\u002F\"\n    - \"scripts\u002F\"\n  file_write:\n    - \"target_skill_directory\"\n    - \"\u002Ftmp\u002F\"\n  shell:\n    allowed_scripts:\n      - \"scripts\u002Fdiscover_assets.py\"\n      - \"scripts\u002Frender_card.py\"\n      - \"scripts\u002Fvalidate_submission.py\"\n---\n\n# Generate Skill Card\n\n**Skill directory to analyze**: $ARGUMENTS\n\n## Purpose\n\nCreate a draft NVIDIA governance skill card for an existing agent skill. The skill gathers source signals, guides the agent to build a grounded JSON context, renders a deterministic markdown card, and checks that human-review markers were removed before submission.\n\nUse this when:\n- A skill directory already exists and needs a new governance card.\n- A changed skill needs its existing card refreshed.\n- A skill owner is preparing legal\u002Fsafety review material.\n\nDo NOT use for:\n- Explaining, listing, comparing, or discussing skills or skill capabilities.\n- Creating or rewriting the source skill itself.\n- Generating cards for non-skill assets such as models, datasets, containers, or full systems.\n- Signing, publishing, or approving a skill card.\n- Replacing required human legal, safety, or owner review.\n\n## Prerequisites\n\n- Python 3 is available.\n- `jinja2` is installed before running `render_card.py`.\n- The target path is a skill directory containing `SKILL.md` or `skill.md`.\n- The agent can write a temporary context JSON file and the rendered card output.\n- Runtime permissions allow reads from `target_skill_directory` plus this skill's `references\u002F` and `scripts\u002F`, writes only to the target skill directory or `\u002Ftmp\u002F`, and shell execution only for the three scripts listed below.\n\n## Instructions\n\n1. First, read this `SKILL.md` completely before running any script.\n2. Resolve the target skill directory from `$ARGUMENTS`; if omitted, use the current working directory.\n3. Stay within the declared permission scope. Do not read `.env`, credential files, hidden auth folders, or unrelated repo files; do not write outside the target skill directory or `\u002Ftmp\u002F`.\n4. Run `scripts\u002Fdiscover_assets.py` against the target. Use the structured signal summary first; if output is truncated, read only targeted files or small excerpts.\n5. Build a context JSON file from the structured signal summary first, then from extracted file contents only when needed.\n6. Populate `credential_requirements` with only two fields: `requires_api_key_or_credential` and `credential_types`. Ground the classification in SKILL.md prose documentation — not script inspection alone. For `credential_types`, use the controlled vocabulary in `references\u002Fstyle-guide.md`. Never include credential values, assignments, or raw environment variable names.\n7. Follow `references\u002Fstyle-guide.md` for every context field. Use `HUMAN-REQUIRED` only when no source supports a truthful value.\n8. Render the card with `scripts\u002Frender_card.py` and fix any schema errors before proceeding.\n9. Review the card manually, remove resolved VERIFY and SELECT markers, then run `scripts\u002Fvalidate_submission.py`.\n10. Before finishing, confirm the rendered card has no unrendered `{{ ... }}` or `{% ... %}` template fragments.\n\n## Available Scripts\n\n| Script | Purpose | Arguments |\n| --- | --- | --- |\n| `scripts\u002Fdiscover_assets.py` | Extracts skill files, repo signals, style guide, and template into one discovery report. | `\u003Cskill_directory>` |\n| `scripts\u002Frender_card.py` | Validates context JSON and renders the skill card from the Jinja template. | `--context \u003Ccontext.json> --template \u003Cskill-card.md.j2> --out \u003Coutput.md>` |\n| `scripts\u002Fvalidate_submission.py` | Fails if the rendered card still contains VERIFY or SELECT review markers. | `\u003Crendered-card.md>` |\n\n## Examples\n\nDiscover signals for a target skill:\n\n```text\nrun_script(\"scripts\u002Fdiscover_assets.py\", args=[\"\u002Fpath\u002Fto\u002Ftarget-skill\"])\n```\n\nRender a card from the completed context:\n\n```text\nrun_script(\n  \"scripts\u002Frender_card.py\",\n  args=[\n    \"--context\", \"\u002Ftmp\u002Ftarget-skill-context.json\",\n    \"--template\", \"references\u002Fskill-card.md.j2\",\n    \"--out\", \"\u002Fpath\u002Fto\u002Ftarget-skill\u002Ftarget-skill-card.md\"\n  ]\n)\n```\n\nValidate the reviewed card before submission:\n\n```text\nrun_script(\"scripts\u002Fvalidate_submission.py\", args=[\"\u002Fpath\u002Fto\u002Ftarget-skill\u002Ftarget-skill-card.md\"])\n```\n\n## Limitations\n\n- The generated card is a draft and must be reviewed by a human owner.\n- Discovery is limited to local files and repo metadata visible from the target path.\n- The renderer validates required context shape, not the legal or safety correctness of field values.\n- Canned limitation and risk catalogs are starting points; remove entries that do not apply.\n\n## Troubleshooting\n\n| Error | Cause | Solution |\n| --- | --- | --- |\n| `directory not found` | The target path is wrong or not mounted in the workspace. | Re-run discovery with the absolute path to the skill directory. |\n| `jinja2 not installed` | The renderer dependency is missing. | Install `jinja2`, then re-run `render_card.py`. |\n| `Context validation failed` | Required fields are missing or typed incorrectly. | Fix the context JSON using `references\u002Fstyle-guide.md`. |\n| Unresolved marker failure | VERIFY or SELECT markers remain after review. | Confirm each marked field, prune catalog entries, then re-run `validate_submission.py`. |\n\n## Files in this skill\n\n- `SKILL.md` - this file (orchestration)\n- `references\u002Fstyle-guide.md` - per-context-field guidance\n- `references\u002Fskill-card.md.j2` - exact card layout\n- `references\u002FSkill Card Generator License.txt` - license text for this skill package\n- `references\u002Fcatalog\u002Flimitations.json` - canned technical-limitations catalog\n- `references\u002Fcatalog\u002Frisks.json` - canned risk-management catalog\n- `scripts\u002Fdiscover_assets.py` - discovery and signal extraction\n- `scripts\u002Frender_card.py` - Jinja renderer with context validation\n- `scripts\u002Fvalidate_submission.py` - pre-submission marker validator\n",{"data":35,"body":54},{"name":4,"description":6,"license":26,"compatibility":36,"metadata":37,"permissions":42},"Any agent that can run Python scripts and write files",{"author":38,"tags":39,"domain":18},"Trustworthy AI Projects \u003Ctrustworthyaiprojects@nvidia.com>",[40,14,18,41],"skill-card","trustworthy-ai",{"file_read":43,"file_write":47,"shell":49},[44,45,46],"target_skill_directory","references\u002F","scripts\u002F",[44,48],"\u002Ftmp\u002F",{"allowed_scripts":50},[51,52,53],"scripts\u002Fdiscover_assets.py","scripts\u002Frender_card.py","scripts\u002Fvalidate_submission.py",{"type":55,"children":56},"root",[57,66,78,85,90,95,115,120,148,154,240,246,417,423,530,536,541,553,558,567,572,581,587,610,616,753,759],{"type":58,"tag":59,"props":60,"children":62},"element","h1",{"id":61},"generate-skill-card",[63],{"type":64,"value":65},"text","Generate Skill Card",{"type":58,"tag":67,"props":68,"children":69},"p",{},[70,76],{"type":58,"tag":71,"props":72,"children":73},"strong",{},[74],{"type":64,"value":75},"Skill directory to analyze",{"type":64,"value":77},": $ARGUMENTS",{"type":58,"tag":79,"props":80,"children":82},"h2",{"id":81},"purpose",[83],{"type":64,"value":84},"Purpose",{"type":58,"tag":67,"props":86,"children":87},{},[88],{"type":64,"value":89},"Create a draft NVIDIA governance skill card for an existing agent skill. The skill gathers source signals, guides the agent to build a grounded JSON context, renders a deterministic markdown card, and checks that human-review markers were removed before submission.",{"type":58,"tag":67,"props":91,"children":92},{},[93],{"type":64,"value":94},"Use this when:",{"type":58,"tag":96,"props":97,"children":98},"ul",{},[99,105,110],{"type":58,"tag":100,"props":101,"children":102},"li",{},[103],{"type":64,"value":104},"A skill directory already exists and needs a new governance card.",{"type":58,"tag":100,"props":106,"children":107},{},[108],{"type":64,"value":109},"A changed skill needs its existing card refreshed.",{"type":58,"tag":100,"props":111,"children":112},{},[113],{"type":64,"value":114},"A skill owner is preparing legal\u002Fsafety review material.",{"type":58,"tag":67,"props":116,"children":117},{},[118],{"type":64,"value":119},"Do NOT use for:",{"type":58,"tag":96,"props":121,"children":122},{},[123,128,133,138,143],{"type":58,"tag":100,"props":124,"children":125},{},[126],{"type":64,"value":127},"Explaining, listing, comparing, or discussing skills or skill capabilities.",{"type":58,"tag":100,"props":129,"children":130},{},[131],{"type":64,"value":132},"Creating or rewriting the source skill itself.",{"type":58,"tag":100,"props":134,"children":135},{},[136],{"type":64,"value":137},"Generating cards for non-skill assets such as models, datasets, containers, or full systems.",{"type":58,"tag":100,"props":139,"children":140},{},[141],{"type":64,"value":142},"Signing, publishing, or approving a skill card.",{"type":58,"tag":100,"props":144,"children":145},{},[146],{"type":64,"value":147},"Replacing required human legal, safety, or owner review.",{"type":58,"tag":79,"props":149,"children":151},{"id":150},"prerequisites",[152],{"type":64,"value":153},"Prerequisites",{"type":58,"tag":96,"props":155,"children":156},{},[157,162,182,202,207],{"type":58,"tag":100,"props":158,"children":159},{},[160],{"type":64,"value":161},"Python 3 is available.",{"type":58,"tag":100,"props":163,"children":164},{},[165,172,174,180],{"type":58,"tag":166,"props":167,"children":169},"code",{"className":168},[],[170],{"type":64,"value":171},"jinja2",{"type":64,"value":173}," is installed before running ",{"type":58,"tag":166,"props":175,"children":177},{"className":176},[],[178],{"type":64,"value":179},"render_card.py",{"type":64,"value":181},".",{"type":58,"tag":100,"props":183,"children":184},{},[185,187,193,195,201],{"type":64,"value":186},"The target path is a skill directory containing ",{"type":58,"tag":166,"props":188,"children":190},{"className":189},[],[191],{"type":64,"value":192},"SKILL.md",{"type":64,"value":194}," or ",{"type":58,"tag":166,"props":196,"children":198},{"className":197},[],[199],{"type":64,"value":200},"skill.md",{"type":64,"value":181},{"type":58,"tag":100,"props":203,"children":204},{},[205],{"type":64,"value":206},"The agent can write a temporary context JSON file and the rendered card output.",{"type":58,"tag":100,"props":208,"children":209},{},[210,212,217,219,224,226,231,233,238],{"type":64,"value":211},"Runtime permissions allow reads from ",{"type":58,"tag":166,"props":213,"children":215},{"className":214},[],[216],{"type":64,"value":44},{"type":64,"value":218}," plus this skill's ",{"type":58,"tag":166,"props":220,"children":222},{"className":221},[],[223],{"type":64,"value":45},{"type":64,"value":225}," and ",{"type":58,"tag":166,"props":227,"children":229},{"className":228},[],[230],{"type":64,"value":46},{"type":64,"value":232},", writes only to the target skill directory or ",{"type":58,"tag":166,"props":234,"children":236},{"className":235},[],[237],{"type":64,"value":48},{"type":64,"value":239},", and shell execution only for the three scripts listed below.",{"type":58,"tag":79,"props":241,"children":243},{"id":242},"instructions",[244],{"type":64,"value":245},"Instructions",{"type":58,"tag":247,"props":248,"children":249},"ol",{},[250,262,275,294,306,311,354,374,386,397],{"type":58,"tag":100,"props":251,"children":252},{},[253,255,260],{"type":64,"value":254},"First, read this ",{"type":58,"tag":166,"props":256,"children":258},{"className":257},[],[259],{"type":64,"value":192},{"type":64,"value":261}," completely before running any script.",{"type":58,"tag":100,"props":263,"children":264},{},[265,267,273],{"type":64,"value":266},"Resolve the target skill directory from ",{"type":58,"tag":166,"props":268,"children":270},{"className":269},[],[271],{"type":64,"value":272},"$ARGUMENTS",{"type":64,"value":274},"; if omitted, use the current working directory.",{"type":58,"tag":100,"props":276,"children":277},{},[278,280,286,288,293],{"type":64,"value":279},"Stay within the declared permission scope. Do not read ",{"type":58,"tag":166,"props":281,"children":283},{"className":282},[],[284],{"type":64,"value":285},".env",{"type":64,"value":287},", credential files, hidden auth folders, or unrelated repo files; do not write outside the target skill directory or ",{"type":58,"tag":166,"props":289,"children":291},{"className":290},[],[292],{"type":64,"value":48},{"type":64,"value":181},{"type":58,"tag":100,"props":295,"children":296},{},[297,299,304],{"type":64,"value":298},"Run ",{"type":58,"tag":166,"props":300,"children":302},{"className":301},[],[303],{"type":64,"value":51},{"type":64,"value":305}," against the target. Use the structured signal summary first; if output is truncated, read only targeted files or small excerpts.",{"type":58,"tag":100,"props":307,"children":308},{},[309],{"type":64,"value":310},"Build a context JSON file from the structured signal summary first, then from extracted file contents only when needed.",{"type":58,"tag":100,"props":312,"children":313},{},[314,316,322,324,330,331,337,339,344,346,352],{"type":64,"value":315},"Populate ",{"type":58,"tag":166,"props":317,"children":319},{"className":318},[],[320],{"type":64,"value":321},"credential_requirements",{"type":64,"value":323}," with only two fields: ",{"type":58,"tag":166,"props":325,"children":327},{"className":326},[],[328],{"type":64,"value":329},"requires_api_key_or_credential",{"type":64,"value":225},{"type":58,"tag":166,"props":332,"children":334},{"className":333},[],[335],{"type":64,"value":336},"credential_types",{"type":64,"value":338},". Ground the classification in SKILL.md prose documentation — not script inspection alone. For ",{"type":58,"tag":166,"props":340,"children":342},{"className":341},[],[343],{"type":64,"value":336},{"type":64,"value":345},", use the controlled vocabulary in ",{"type":58,"tag":166,"props":347,"children":349},{"className":348},[],[350],{"type":64,"value":351},"references\u002Fstyle-guide.md",{"type":64,"value":353},". Never include credential values, assignments, or raw environment variable names.",{"type":58,"tag":100,"props":355,"children":356},{},[357,359,364,366,372],{"type":64,"value":358},"Follow ",{"type":58,"tag":166,"props":360,"children":362},{"className":361},[],[363],{"type":64,"value":351},{"type":64,"value":365}," for every context field. Use ",{"type":58,"tag":166,"props":367,"children":369},{"className":368},[],[370],{"type":64,"value":371},"HUMAN-REQUIRED",{"type":64,"value":373}," only when no source supports a truthful value.",{"type":58,"tag":100,"props":375,"children":376},{},[377,379,384],{"type":64,"value":378},"Render the card with ",{"type":58,"tag":166,"props":380,"children":382},{"className":381},[],[383],{"type":64,"value":52},{"type":64,"value":385}," and fix any schema errors before proceeding.",{"type":58,"tag":100,"props":387,"children":388},{},[389,391,396],{"type":64,"value":390},"Review the card manually, remove resolved VERIFY and SELECT markers, then run ",{"type":58,"tag":166,"props":392,"children":394},{"className":393},[],[395],{"type":64,"value":53},{"type":64,"value":181},{"type":58,"tag":100,"props":398,"children":399},{},[400,402,408,409,415],{"type":64,"value":401},"Before finishing, confirm the rendered card has no unrendered ",{"type":58,"tag":166,"props":403,"children":405},{"className":404},[],[406],{"type":64,"value":407},"{{ ... }}",{"type":64,"value":194},{"type":58,"tag":166,"props":410,"children":412},{"className":411},[],[413],{"type":64,"value":414},"{% ... %}",{"type":64,"value":416}," template fragments.",{"type":58,"tag":79,"props":418,"children":420},{"id":419},"available-scripts",[421],{"type":64,"value":422},"Available Scripts",{"type":58,"tag":424,"props":425,"children":426},"table",{},[427,450],{"type":58,"tag":428,"props":429,"children":430},"thead",{},[431],{"type":58,"tag":432,"props":433,"children":434},"tr",{},[435,441,445],{"type":58,"tag":436,"props":437,"children":438},"th",{},[439],{"type":64,"value":440},"Script",{"type":58,"tag":436,"props":442,"children":443},{},[444],{"type":64,"value":84},{"type":58,"tag":436,"props":446,"children":447},{},[448],{"type":64,"value":449},"Arguments",{"type":58,"tag":451,"props":452,"children":453},"tbody",{},[454,480,505],{"type":58,"tag":432,"props":455,"children":456},{},[457,466,471],{"type":58,"tag":458,"props":459,"children":460},"td",{},[461],{"type":58,"tag":166,"props":462,"children":464},{"className":463},[],[465],{"type":64,"value":51},{"type":58,"tag":458,"props":467,"children":468},{},[469],{"type":64,"value":470},"Extracts skill files, repo signals, style guide, and template into one discovery report.",{"type":58,"tag":458,"props":472,"children":473},{},[474],{"type":58,"tag":166,"props":475,"children":477},{"className":476},[],[478],{"type":64,"value":479},"\u003Cskill_directory>",{"type":58,"tag":432,"props":481,"children":482},{},[483,491,496],{"type":58,"tag":458,"props":484,"children":485},{},[486],{"type":58,"tag":166,"props":487,"children":489},{"className":488},[],[490],{"type":64,"value":52},{"type":58,"tag":458,"props":492,"children":493},{},[494],{"type":64,"value":495},"Validates context JSON and renders the skill card from the Jinja template.",{"type":58,"tag":458,"props":497,"children":498},{},[499],{"type":58,"tag":166,"props":500,"children":502},{"className":501},[],[503],{"type":64,"value":504},"--context \u003Ccontext.json> --template \u003Cskill-card.md.j2> --out \u003Coutput.md>",{"type":58,"tag":432,"props":506,"children":507},{},[508,516,521],{"type":58,"tag":458,"props":509,"children":510},{},[511],{"type":58,"tag":166,"props":512,"children":514},{"className":513},[],[515],{"type":64,"value":53},{"type":58,"tag":458,"props":517,"children":518},{},[519],{"type":64,"value":520},"Fails if the rendered card still contains VERIFY or SELECT review markers.",{"type":58,"tag":458,"props":522,"children":523},{},[524],{"type":58,"tag":166,"props":525,"children":527},{"className":526},[],[528],{"type":64,"value":529},"\u003Crendered-card.md>",{"type":58,"tag":79,"props":531,"children":533},{"id":532},"examples",[534],{"type":64,"value":535},"Examples",{"type":58,"tag":67,"props":537,"children":538},{},[539],{"type":64,"value":540},"Discover signals for a target skill:",{"type":58,"tag":542,"props":543,"children":548},"pre",{"className":544,"code":546,"language":64,"meta":547},[545],"language-text","run_script(\"scripts\u002Fdiscover_assets.py\", args=[\"\u002Fpath\u002Fto\u002Ftarget-skill\"])\n","",[549],{"type":58,"tag":166,"props":550,"children":551},{"__ignoreMap":547},[552],{"type":64,"value":546},{"type":58,"tag":67,"props":554,"children":555},{},[556],{"type":64,"value":557},"Render a card from the completed context:",{"type":58,"tag":542,"props":559,"children":562},{"className":560,"code":561,"language":64,"meta":547},[545],"run_script(\n  \"scripts\u002Frender_card.py\",\n  args=[\n    \"--context\", \"\u002Ftmp\u002Ftarget-skill-context.json\",\n    \"--template\", \"references\u002Fskill-card.md.j2\",\n    \"--out\", \"\u002Fpath\u002Fto\u002Ftarget-skill\u002Ftarget-skill-card.md\"\n  ]\n)\n",[563],{"type":58,"tag":166,"props":564,"children":565},{"__ignoreMap":547},[566],{"type":64,"value":561},{"type":58,"tag":67,"props":568,"children":569},{},[570],{"type":64,"value":571},"Validate the reviewed card before submission:",{"type":58,"tag":542,"props":573,"children":576},{"className":574,"code":575,"language":64,"meta":547},[545],"run_script(\"scripts\u002Fvalidate_submission.py\", args=[\"\u002Fpath\u002Fto\u002Ftarget-skill\u002Ftarget-skill-card.md\"])\n",[577],{"type":58,"tag":166,"props":578,"children":579},{"__ignoreMap":547},[580],{"type":64,"value":575},{"type":58,"tag":79,"props":582,"children":584},{"id":583},"limitations",[585],{"type":64,"value":586},"Limitations",{"type":58,"tag":96,"props":588,"children":589},{},[590,595,600,605],{"type":58,"tag":100,"props":591,"children":592},{},[593],{"type":64,"value":594},"The generated card is a draft and must be reviewed by a human owner.",{"type":58,"tag":100,"props":596,"children":597},{},[598],{"type":64,"value":599},"Discovery is limited to local files and repo metadata visible from the target path.",{"type":58,"tag":100,"props":601,"children":602},{},[603],{"type":64,"value":604},"The renderer validates required context shape, not the legal or safety correctness of field values.",{"type":58,"tag":100,"props":606,"children":607},{},[608],{"type":64,"value":609},"Canned limitation and risk catalogs are starting points; remove entries that do not apply.",{"type":58,"tag":79,"props":611,"children":613},{"id":612},"troubleshooting",[614],{"type":64,"value":615},"Troubleshooting",{"type":58,"tag":424,"props":617,"children":618},{},[619,640],{"type":58,"tag":428,"props":620,"children":621},{},[622],{"type":58,"tag":432,"props":623,"children":624},{},[625,630,635],{"type":58,"tag":436,"props":626,"children":627},{},[628],{"type":64,"value":629},"Error",{"type":58,"tag":436,"props":631,"children":632},{},[633],{"type":64,"value":634},"Cause",{"type":58,"tag":436,"props":636,"children":637},{},[638],{"type":64,"value":639},"Solution",{"type":58,"tag":451,"props":641,"children":642},{},[643,665,700,728],{"type":58,"tag":432,"props":644,"children":645},{},[646,655,660],{"type":58,"tag":458,"props":647,"children":648},{},[649],{"type":58,"tag":166,"props":650,"children":652},{"className":651},[],[653],{"type":64,"value":654},"directory not found",{"type":58,"tag":458,"props":656,"children":657},{},[658],{"type":64,"value":659},"The target path is wrong or not mounted in the workspace.",{"type":58,"tag":458,"props":661,"children":662},{},[663],{"type":64,"value":664},"Re-run discovery with the absolute path to the skill directory.",{"type":58,"tag":432,"props":666,"children":667},{},[668,677,682],{"type":58,"tag":458,"props":669,"children":670},{},[671],{"type":58,"tag":166,"props":672,"children":674},{"className":673},[],[675],{"type":64,"value":676},"jinja2 not installed",{"type":58,"tag":458,"props":678,"children":679},{},[680],{"type":64,"value":681},"The renderer dependency is missing.",{"type":58,"tag":458,"props":683,"children":684},{},[685,687,692,694,699],{"type":64,"value":686},"Install ",{"type":58,"tag":166,"props":688,"children":690},{"className":689},[],[691],{"type":64,"value":171},{"type":64,"value":693},", then re-run ",{"type":58,"tag":166,"props":695,"children":697},{"className":696},[],[698],{"type":64,"value":179},{"type":64,"value":181},{"type":58,"tag":432,"props":701,"children":702},{},[703,712,717],{"type":58,"tag":458,"props":704,"children":705},{},[706],{"type":58,"tag":166,"props":707,"children":709},{"className":708},[],[710],{"type":64,"value":711},"Context validation failed",{"type":58,"tag":458,"props":713,"children":714},{},[715],{"type":64,"value":716},"Required fields are missing or typed incorrectly.",{"type":58,"tag":458,"props":718,"children":719},{},[720,722,727],{"type":64,"value":721},"Fix the context JSON using ",{"type":58,"tag":166,"props":723,"children":725},{"className":724},[],[726],{"type":64,"value":351},{"type":64,"value":181},{"type":58,"tag":432,"props":729,"children":730},{},[731,736,741],{"type":58,"tag":458,"props":732,"children":733},{},[734],{"type":64,"value":735},"Unresolved marker failure",{"type":58,"tag":458,"props":737,"children":738},{},[739],{"type":64,"value":740},"VERIFY or SELECT markers remain after review.",{"type":58,"tag":458,"props":742,"children":743},{},[744,746,752],{"type":64,"value":745},"Confirm each marked field, prune catalog entries, then re-run ",{"type":58,"tag":166,"props":747,"children":749},{"className":748},[],[750],{"type":64,"value":751},"validate_submission.py",{"type":64,"value":181},{"type":58,"tag":79,"props":754,"children":756},{"id":755},"files-in-this-skill",[757],{"type":64,"value":758},"Files in this skill",{"type":58,"tag":96,"props":760,"children":761},{},[762,772,782,793,804,815,826,836,846],{"type":58,"tag":100,"props":763,"children":764},{},[765,770],{"type":58,"tag":166,"props":766,"children":768},{"className":767},[],[769],{"type":64,"value":192},{"type":64,"value":771}," - this file (orchestration)",{"type":58,"tag":100,"props":773,"children":774},{},[775,780],{"type":58,"tag":166,"props":776,"children":778},{"className":777},[],[779],{"type":64,"value":351},{"type":64,"value":781}," - per-context-field guidance",{"type":58,"tag":100,"props":783,"children":784},{},[785,791],{"type":58,"tag":166,"props":786,"children":788},{"className":787},[],[789],{"type":64,"value":790},"references\u002Fskill-card.md.j2",{"type":64,"value":792}," - exact card layout",{"type":58,"tag":100,"props":794,"children":795},{},[796,802],{"type":58,"tag":166,"props":797,"children":799},{"className":798},[],[800],{"type":64,"value":801},"references\u002FSkill Card Generator License.txt",{"type":64,"value":803}," - license text for this skill package",{"type":58,"tag":100,"props":805,"children":806},{},[807,813],{"type":58,"tag":166,"props":808,"children":810},{"className":809},[],[811],{"type":64,"value":812},"references\u002Fcatalog\u002Flimitations.json",{"type":64,"value":814}," - canned technical-limitations catalog",{"type":58,"tag":100,"props":816,"children":817},{},[818,824],{"type":58,"tag":166,"props":819,"children":821},{"className":820},[],[822],{"type":64,"value":823},"references\u002Fcatalog\u002Frisks.json",{"type":64,"value":825}," - canned risk-management catalog",{"type":58,"tag":100,"props":827,"children":828},{},[829,834],{"type":58,"tag":166,"props":830,"children":832},{"className":831},[],[833],{"type":64,"value":51},{"type":64,"value":835}," - discovery and signal extraction",{"type":58,"tag":100,"props":837,"children":838},{},[839,844],{"type":58,"tag":166,"props":840,"children":842},{"className":841},[],[843],{"type":64,"value":52},{"type":64,"value":845}," - Jinja renderer with context validation",{"type":58,"tag":100,"props":847,"children":848},{},[849,854],{"type":58,"tag":166,"props":850,"children":852},{"className":851},[],[853],{"type":64,"value":53},{"type":64,"value":855}," - pre-submission marker validator",{"items":857,"total":961},[858,875,889,903,915,932,947],{"slug":859,"name":859,"fn":860,"description":861,"org":862,"tags":863,"stars":23,"repoUrl":24,"updatedAt":874},"accelerated-computing-cudf","accelerate data processing with cuDF","Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV\u002FParquet I\u002FO, nullable semantics, and multi-GPU DataFrame workloads.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[864,867,870,871],{"name":865,"slug":866,"type":15},"Data Analysis","data-analysis",{"name":868,"slug":869,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":872,"slug":873,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":876,"name":876,"fn":877,"description":878,"org":879,"tags":880,"stars":23,"repoUrl":24,"updatedAt":888},"aiq-deploy","deploy and manage NVIDIA AI-Q infrastructure","Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[881,884,887],{"name":882,"slug":883,"type":15},"Deployment","deployment",{"name":885,"slug":886,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":890,"name":890,"fn":891,"description":892,"org":893,"tags":894,"stars":23,"repoUrl":24,"updatedAt":902},"aiq-research","conduct deep research with AI-Q","Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[895,898,899],{"name":896,"slug":897,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":900,"slug":901,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":904,"name":904,"fn":905,"description":906,"org":907,"tags":908,"stars":23,"repoUrl":24,"updatedAt":914},"amc-run-sample-calibration","run AMC sample dataset calibration","Run end-to-end calibration on the shipped sample dataset (sdg_08_2_sample_data_010926.zip) against a running AMC microservice. Use when user says 'test sample dataset', 'run sample calibration', 'verify AMC install', or 'launch and test'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[909,910,911],{"name":865,"slug":866,"type":15},{"name":9,"slug":8,"type":15},{"name":912,"slug":913,"type":15},"Testing","testing","2026-07-17T05:29:03.913266",{"slug":916,"name":916,"fn":917,"description":918,"org":919,"tags":920,"stars":23,"repoUrl":24,"updatedAt":931},"amc-run-video-calibration","calibrate video datasets with AutoMagicCalib","Calibrate a new dataset from pre-recorded video files via the AutoMagicCalib REST API. Use when user has local MP4s and says 'calibrate my videos', 'run AMC on these videos', or similar. For RTSP\u002Flive streams, use amc-run-rtsp-calibration instead.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[921,924,927,928],{"name":922,"slug":923,"type":15},"Automation","automation",{"name":925,"slug":926,"type":15},"Imaging","imaging",{"name":9,"slug":8,"type":15},{"name":929,"slug":930,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":933,"name":933,"fn":934,"description":935,"org":936,"tags":937,"stars":23,"repoUrl":24,"updatedAt":946},"amc-setup-calibration-stack","deploy AutoMagicCalib microservice with Docker","Launch AutoMagicCalib microservice and web UI from NGC release images via Docker Compose. Use when user says 'deploy auto calibration', 'launch auto calibration', 'launch AMC', 'start MS+UI', or 'set up auto-magic-calib'. Requires NGC API key.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[938,939,942,943],{"name":882,"slug":883,"type":15},{"name":940,"slug":941,"type":15},"Docker","docker",{"name":9,"slug":8,"type":15},{"name":944,"slug":945,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":948,"name":948,"fn":949,"description":950,"org":951,"tags":952,"stars":23,"repoUrl":24,"updatedAt":960},"cudaq-guide","develop quantum applications with CUDA-Q","CUDA-Q onboarding guide for installation, test programs, GPU simulation, QPU hardware, and quantum applications.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[953,954,957],{"name":9,"slug":8,"type":15},{"name":955,"slug":956,"type":15},"Quantum Computing","quantum-computing",{"name":958,"slug":959,"type":15},"Simulation","simulation","2026-07-14T05:26:58.898253",305,{"items":963,"total":1112},[964,980,996,1007,1019,1033,1046,1060,1071,1080,1094,1103],{"slug":965,"name":965,"fn":966,"description":967,"org":968,"tags":969,"stars":977,"repoUrl":978,"updatedAt":979},"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},[970,971,974],{"name":17,"slug":18,"type":15},{"name":972,"slug":973,"type":15},"MCP","mcp",{"name":975,"slug":976,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":981,"name":981,"fn":982,"description":983,"org":984,"tags":985,"stars":993,"repoUrl":994,"updatedAt":995},"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},[986,989,990],{"name":987,"slug":988,"type":15},"Containers","containers",{"name":882,"slug":883,"type":15},{"name":991,"slug":992,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":997,"name":997,"fn":998,"description":999,"org":1000,"tags":1001,"stars":993,"repoUrl":994,"updatedAt":1006},"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},[1002,1005],{"name":1003,"slug":1004,"type":15},"CI\u002FCD","ci-cd",{"name":882,"slug":883,"type":15},"2026-07-14T05:25:59.97109",{"slug":1008,"name":1008,"fn":1009,"description":1010,"org":1011,"tags":1012,"stars":993,"repoUrl":994,"updatedAt":1018},"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},[1013,1014,1015],{"name":1003,"slug":1004,"type":15},{"name":882,"slug":883,"type":15},{"name":1016,"slug":1017,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1020,"name":1020,"fn":1021,"description":1022,"org":1023,"tags":1024,"stars":993,"repoUrl":994,"updatedAt":1032},"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},[1025,1028,1029],{"name":1026,"slug":1027,"type":15},"Debugging","debugging",{"name":1016,"slug":1017,"type":15},{"name":1030,"slug":1031,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1034,"name":1034,"fn":1035,"description":1036,"org":1037,"tags":1038,"stars":993,"repoUrl":994,"updatedAt":1045},"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},[1039,1042],{"name":1040,"slug":1041,"type":15},"Best Practices","best-practices",{"name":1043,"slug":1044,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1047,"name":1047,"fn":1048,"description":1049,"org":1050,"tags":1051,"stars":993,"repoUrl":994,"updatedAt":1059},"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},[1052,1055,1058],{"name":1053,"slug":1054,"type":15},"Machine Learning","machine-learning",{"name":1056,"slug":1057,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1061,"name":1061,"fn":1062,"description":1063,"org":1064,"tags":1065,"stars":993,"repoUrl":994,"updatedAt":1070},"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},[1066,1069],{"name":1067,"slug":1068,"type":15},"QA","qa",{"name":912,"slug":913,"type":15},"2026-07-14T05:25:53.673039",{"slug":1072,"name":1072,"fn":1073,"description":1074,"org":1075,"tags":1076,"stars":993,"repoUrl":994,"updatedAt":1079},"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},[1077,1078],{"name":882,"slug":883,"type":15},{"name":885,"slug":886,"type":15},"2026-07-14T05:25:49.362534",{"slug":1081,"name":1081,"fn":1082,"description":1083,"org":1084,"tags":1085,"stars":993,"repoUrl":994,"updatedAt":1093},"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},[1086,1089,1090],{"name":1087,"slug":1088,"type":15},"Code Review","code-review",{"name":1016,"slug":1017,"type":15},{"name":1091,"slug":1092,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1095,"name":1095,"fn":1096,"description":1097,"org":1098,"tags":1099,"stars":993,"repoUrl":994,"updatedAt":1102},"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},[1100,1101],{"name":1067,"slug":1068,"type":15},{"name":912,"slug":913,"type":15},"2026-07-14T05:25:54.928983",{"slug":1104,"name":1104,"fn":1105,"description":1106,"org":1107,"tags":1108,"stars":993,"repoUrl":994,"updatedAt":1111},"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},[1109,1110],{"name":922,"slug":923,"type":15},{"name":1003,"slug":1004,"type":15},"2026-07-30T05:29:03.275638",496]