[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-physical-ai-neural-reconstruction":3,"mdc--7qovf9-key":34,"related-org-nvidia-physical-ai-neural-reconstruction":3123,"related-repo-nvidia-physical-ai-neural-reconstruction":3284},{"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},"physical-ai-neural-reconstruction","perform neural reconstruction for physical AI","Router for NVIDIA NuRec\u002FNRE: USDZ rendering, NCore conversion, 3DGS, gRPC sensor sim, PhysicalAI HF datasets. Do NOT use for SimReady or infra setup.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,17,20],{"name":13,"slug":14,"type":15},"3D","3d","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Simulation","simulation",{"name":21,"slug":22,"type":15},"Computer Vision","computer-vision",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-17T05:29:04.920397","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\u002Fphysical-ai-neural-reconstruction","---\nname: physical-ai-neural-reconstruction\ndescription: \"Router for NVIDIA NuRec\u002FNRE: USDZ rendering, NCore conversion, 3DGS, gRPC sensor sim, PhysicalAI HF datasets. Do NOT use for SimReady or infra setup.\"\nlicense: Apache-2.0\nversion: \"0.3.0\"\ntools:\n  - Read\n  - Shell\ncompatibility: >-\n  Router skill; downstream sibling skills require Docker, NVIDIA Container\n  Toolkit, GPU, NGC API key, Hugging Face token with PhysicalAI gated\n  licenses accepted, Python 3.10+, and `huggingface_hub`. Optional:\n  CARLA \u002F Isaac Sim 5.1 \u002F AlpaSim for simulator integration over\n  `serve-grpc`.\nmetadata:\n  author: NVIDIA Physical AI\n  tags:\n    - physical-ai\n    - nurec\n    - neural-reconstruction\n  upstream:\n    repo: https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills\n    branch: main\n    skills_dir: .agents\u002Fskills\u002F\n    skills_dir_alias: skills\u002F\n    index_skill: .agents\u002Fskills\u002FSKILL.md\n    index_skill_name: nurec-index\n    sibling_skills:\n      - name: physical-ai-datasets\n        folder: physical-ai-datasets\u002F\n        upstream: https:\u002F\u002Fhuggingface.co\u002Fnvidia\n      - name: ncore\n        folder: ncore\u002F\n        upstream: https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fncore\n      - name: nre\n        folder: nre\u002F\n        upstream: nvcr.io\u002Fnvidia\u002Fnre\u002Fnre\n      - name: asset-harvester\n        folder: asset-harvester\u002F\n        upstream: https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fasset-harvester\n      - name: nurec-fixer\n        folder: nurec-fixer\u002F\n        upstream: https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fharmonizer\n        hf_model: https:\u002F\u002Fhuggingface.co\u002Fnvidia\u002FDiffusionHarmonizer\n  upstream_clone_path: \"${PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT:-$HOME\u002F.physical-ai-skill-hub\u002Fupstreams}\u002Fnurec-skills\"\n  upstream_override_env: NUREC_SKILLS_UPSTREAM_ROOT\n---\n\n# Physical AI Neural Reconstruction (NuRec) Router\n\n## Purpose\n\nThis is a **thin router** for NVIDIA Neural Reconstruction (NuRec)\nrequests. It points at the upstream `nurec-index` skill at\n`https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills` and its five sibling skills\n(`physical-ai-datasets`, `ncore`, `nre`, `asset-harvester`,\n`nurec-fixer`). Use this skill to:\n\n- Identify which upstream sibling skill answers a NuRec question.\n- Locate, clone, or refresh the canonical `nurec-skills` checkout.\n- Order multi-step NuRec workflows (data → conversion → train →\n  render → cleanup) before opening the upstream recipe.\n\nThe canonical recipes (training, rendering, data conversion, dataset\ndownloads, object harvesting, frame cleanup) live in the upstream\nsibling skills. **Never copy or reconstruct their commands here.**\n\n**Do NOT use this skill for:**\n\n- SimReady packaging of CAD or source meshes → use\n  `omniverse-cad-to-simready`.\n- Generic USD performance tuning unrelated to NuRec → use\n  `omniverse-usd-performance-tuning`.\n- AKS \u002F OSMO \u002F NIM Operator infrastructure setup → use\n  `physical-ai-infrastructure-setup-and-resilient-scaling`.\n\n## When to Use\n\nRead this skill **first** whenever a user mentions any of:\n\n`nurec`, `nurec router`, `nurec index`, `neural reconstruction`,\n`neural reconstruction engine`, `NRE`, `3DGUT`, `3DGRT`, `USDZ`,\n`NCore V4`, `sensor sim`, `novel view synthesis`,\n`PhysicalAI-Autonomous-Vehicles-NuRec`, `PhysicalAI-NuRec-PPISP`,\n`Cosmos-Drive-Dreams`, `asset harvester`, `nurec fixer`,\n`DiffusionHarmonizer`, `harmonizer`, `difix`, `difix3d`, `serve-grpc`,\n`render-grpc`, `warm serve-grpc`, `nre thin client`, `batch_render_rgb`,\n`nurec teardown`, \"where do I start with NuRec\", \"which NuRec skill\nshould I use for X?\".\n\nDecide which upstream sibling skill answers the question, fetch it\n(see [Locate and fetch the upstream skills](#locate-and-fetch-the-upstream-skills)),\nthen follow that skill's body.\n\n## Prerequisites\n\nRouter skill itself has no runtime prerequisites beyond `git` for\nfetching the upstream. Downstream sibling skills require:\n\n- **Docker + NVIDIA Container Toolkit + GPU** — for `nre`, `nre-tools`,\n  and `nurec-fixer` containers\n  (`nvcr.io\u002Fnvidia\u002Fnre\u002Fnre`, `nvcr.io\u002Fnvidia\u002Fnre\u002Fnre-tools`,\n  `nvcr.io\u002Fnvidia\u002Fcosmos\u002Fcosmos-predict2-container:1.2`).\n- **NGC API key** (`NGC_API_KEY`) — for pulling NGC containers.\n- **Hugging Face token** (`HF_TOKEN`) with the\n  `nvidia\u002FPhysicalAI-*`, `nvidia\u002FDiffusionHarmonizer`, and\n  `nvidia\u002Fasset-harvester` gated licenses **accepted in advance** on\n  Hugging Face.\n- **Python 3.10+** with `huggingface_hub` installed.\n- **(Optional)** CARLA, Isaac Sim 5.1, or AlpaSim for simulator\n  integration over `serve-grpc`.\n\nVerify secrets safely (do not echo values):\n\n```bash\nhf auth whoami\n[ -n \"${HF_TOKEN:-}\" ]      && echo \"HF_TOKEN length=${#HF_TOKEN}\"      || echo \"HF_TOKEN unset\"\n[ -n \"${NGC_API_KEY:-}\" ]   && echo \"NGC_API_KEY length=${#NGC_API_KEY}\" || echo \"NGC_API_KEY unset\"\n```\n\nSee [`references\u002Fsecrets-handling.md`](references\u002Fsecrets-handling.md)\nfor the bash anti-patterns to avoid.\n\n## What is NuRec?\n\n**NuRec** (NVIDIA Omniverse Neural Reconstruction) takes camera, LiDAR,\nradar, or stereo recordings — typically from a self-driving car or a\nrobot — and turns them into a 3D scene you can re-render from any\nviewpoint. Names that come up a lot:\n\n- **NRE** — \"Neural Reconstruction Engine\". NuRec is the product; NRE\n  is the engine that trains and renders. Both route to the upstream\n  `nre` skill.\n- **USDZ** — the file format of a trained scene. A zip archive that\n  Omniverse, Isaac Sim, and CARLA can open.\n- **NCore V4** — the input format NRE consumes. Raw recordings must be\n  converted to NCore V4 before training.\n- **3DGUT \u002F 3DGRT** — the two 3D Gaussian Splatting flavours used\n  internally by NRE. The default Hydra recipe picks one; most users\n  never set it manually.\n\nA typical NuRec project has three stages:\n\n1. **Get the input** — convert your own recording to NCore V4\n   (`ncore`), or download a pre-converted dataset\n   (`physical-ai-datasets`).\n2. **Train the reconstruction** — feed NCore V4 to NRE; out comes a\n   USDZ (`nre`).\n3. **Render new views** — render images, videos, or LiDAR sweeps from\n   the USDZ (`nre`).\n\nProjects that just want to *use* an existing NVIDIA-published scene\nskip step 2.\n\n## Pick a skill\n\nMatch the user's goal in the left column and open the named upstream\nskill on the right. Arrows mean \"do these in order\".\n\n| I want to… | Upstream skill |\n|------------|----------------|\n| Find or download a NuRec dataset NVIDIA has published | `physical-ai-datasets` |\n| Convert my own camera \u002F LiDAR \u002F radar \u002F depth \u002F stereo recording into NCore V4 | `ncore` |\n| Write a new converter for an unsupported sensor setup (drone, RGB-D, ROS 2 bag, COLMAP, ScanNet++) | `ncore` |\n| Train a 3D reconstruction from an NCore clip | `ncore` → `nre` |\n| Generate the extra inputs NRE needs (segmentation masks, depth, ego mask) | `nre` (uses the `nre-tools` container) |\n| Render a USDZ along the original camera positions | `nre` |\n| Render at full resolution \u002F highest quality | `nre` (see \"Quality presets\") |\n| Render along a shifted trajectory (e.g. car moved 3 m left) | `nre` |\n| Render through a server so CARLA \u002F Isaac Sim \u002F AlpaSim \u002F a custom simulator can ask for frames | `nre` (`serve-grpc`) |\n| Render the same USDZ many times back-to-back from Python with minimal per-call latency | `nre` (warm `serve-grpc` + thin Python client \u002F `batch_render_rgb`) |\n| Render LiDAR sweeps (point clouds) from a USDZ | `nre` (`render-grpc --lidar`) |\n| Skip training and just render a NuRec scene NVIDIA already built | `physical-ai-datasets` → `nre` |\n| Extract individual 3D objects (cars, pedestrians) from a driving clip | `asset-harvester` |\n| Add, remove, or replace cars \u002F pedestrians in a NuRec scene | `asset-harvester` → `nre` |\n| Clean up or harmonize rendered frames (ghosting, floaters, flicker, lighting\u002Fshadows) | `nurec-fixer`, **or** `--enable-difix` inside `nre` for inline rendering |\n| Export the scene as a PLY, mesh, depth maps, ego mask, etc. | `nre` |\n| Upgrade an old USDZ so newer NRE versions load it faster | `nre` (`upgrade-artifact`) |\n| Open a USDZ or PLY in a browser viewer | `nre` (`viewer` \u002F `ply_viewer`) |\n| Measure rendering quality (PSNR, SSIM, LPIPS) against ground truth | `nre` (`eval-rendering-metrics`) |\n| Benchmark different reconstruction methods on the same scenes | `physical-ai-datasets` (`PhysicalAI-NuRec-PPISP`) → `nre` |\n| Train on multiple GPUs or on SLURM | `nre` (Workflow D) |\n\n## Common workflows\n\nSix end-to-end workflows are documented in\n[`references\u002Fworkflows.md`](references\u002Fworkflows.md):\n\n- **A.** Make a NuRec scene from your own recording.\n- **B.** Use a NuRec scene NVIDIA has already trained.\n- **C.** Add, remove, or replace 3D objects in a scene.\n- **D.** Clean up rendered frames.\n- **E.** Benchmark reconstruction quality.\n- **F.** Connect NuRec to a simulator.\n\nOpen that file when the user's task spans more than one sibling skill.\n\n## Sibling skills (upstream)\n\n| Name | Upstream folder | What it does |\n|------|-----------------|--------------|\n| `physical-ai-datasets` | `.agents\u002Fskills\u002Fphysical-ai-datasets\u002F` | Catalog and download recipes for every NVIDIA Physical AI dataset on Hugging Face (driving, robotics, manipulation, NuRec scenes, benchmarks). |\n| `ncore` | `.agents\u002Fskills\u002Fncore\u002F` | Converts any sensor recording to NCore V4 (the format NRE needs). Also covers writing a new converter. |\n| `nre` | `.agents\u002Fskills\u002Fnre\u002F` | The Neural Reconstruction Engine itself. Trains, renders (locally, via warm `serve-grpc` + thin Python client \u002F `batch_render_rgb`, or to an external simulator), exports meshes \u002F point clouds \u002F depth, edits actors, evaluates quality. |\n| `asset-harvester` | `.agents\u002Fskills\u002Fasset-harvester\u002F` | Open-source Apache-2.0 pipeline that extracts individual 3D objects from sparse views in a driving clip and saves them as `.ply` Gaussian splats with metadata. |\n| `nurec-fixer` | `.agents\u002Fskills\u002Fnurec-fixer\u002F` | Standalone NVIDIA **DiffusionHarmonizer** workflow — public successor to the older Fixer \u002F Difix3D+ recipes — that cleans rendered frames, harmonizes inserted actors, evaluates PSNR\u002FLPIPS, and optionally fine-tunes the model. |\n\nFor naming overlaps (NRE vs Fixer, ncore vs nre, AV-NuRec vs\nCosmos-Drive-Dreams, NuRec vs SimReady) see\n[`references\u002Fmix-ups.md`](references\u002Fmix-ups.md).\n\n## Locate and fetch the upstream skills\n\nQuick recipe (full version in\n[`references\u002Fupstream-fetch.md`](references\u002Fupstream-fetch.md)):\n\n```bash\nUPSTREAM_ROOT=\"${NUREC_SKILLS_UPSTREAM_ROOT:-${PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT:-$HOME\u002F.physical-ai-skill-hub\u002Fupstreams}}\"\nmkdir -p \"$UPSTREAM_ROOT\"\nif [ -d \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.git\" ]; then\n  git -C \"$UPSTREAM_ROOT\u002Fnurec-skills\" fetch --tags\n  git -C \"$UPSTREAM_ROOT\u002Fnurec-skills\" checkout main\n  git -C \"$UPSTREAM_ROOT\u002Fnurec-skills\" pull --ff-only\nelse\n  git clone --depth 1 https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills.git \\\n    \"$UPSTREAM_ROOT\u002Fnurec-skills\"\nfi\ntest -f \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.agents\u002Fskills\u002FSKILL.md\"\n```\n\nThen read the upstream skill before running any mutating command:\n\n```bash\ncat \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.agents\u002Fskills\u002FSKILL.md\"          # router\ncat \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.agents\u002Fskills\u002F\u003Cfolder>\u002FSKILL.md\" # sibling\n```\n\nLocal lookup order (try in order before the upstream clone):\n\n1. `.agents\u002Fskills\u002F\u003Cname>\u002FSKILL.md` (Cursor, Codex, NemoClaw)\n2. `.claude\u002Fskills\u002F\u003Cname>\u002FSKILL.md` (Claude Code)\n3. `.cursor\u002Fskills\u002F\u003Cname>\u002FSKILL.md` (project-scoped)\n4. `~\u002F.cursor\u002Fskills\u002F\u003Cname>\u002FSKILL.md` (personal skills)\n\n## Hard Rules\n\n- Router only — do not duplicate upstream NuRec recipes here. Read\n  the upstream sibling skill body before running any mutating command.\n- Refer to sibling skills by their `name:` (e.g. `nre`), not by repo\n  path. Folder layouts can change; the name is portable.\n- Clone or refresh `https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills` under the\n  shared upstream root\n  (`${NUREC_SKILLS_UPSTREAM_ROOT:-${PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT:-$HOME\u002F.physical-ai-skill-hub\u002Fupstreams}}\u002Fnurec-skills`).\n  Do not scan broad developer workspaces such as `~\u002FCodes` or reuse\n  unrelated old clones.\n- `physical-ai-datasets` covers gated Hugging Face datasets. Do not\n  bypass dataset license terms; the user must accept the\n  `PhysicalAI-*` gated licenses on Hugging Face and provide a token\n  before downloading.\n- Asset Harvester runs **before** packaging into a USDZ. Do not call\n  `nre`'s `export-external-assets` on hand-rolled `.ply` files unless\n  the user explicitly asks to skip Asset Harvester.\n- For artifact cleanup, prefer the built-in `--enable-difix` path in\n  `nre`. Route to the standalone `nurec-fixer` only when the user\n  needs the public code\u002Fmodel card, paired evaluation, fine-tuning,\n  or fixes on previously rendered frames.\n- Do not invent NRE \u002F NCore \u002F DiffusionHarmonizer commands from\n  memory. Re-read the upstream sibling skill — versions move fast\n  (NRE `release_26.04` is the current pinned tag).\n- This router does not deploy infrastructure. Route AKS \u002F OSMO \u002F\n  NIM Operator setup to\n  `physical-ai-infrastructure-setup-and-resilient-scaling`.\n\n## Limitations\n\n- **Router only.** This skill never executes mutating NuRec commands.\n  All training, rendering, conversion, and harmonization happens in\n  upstream sibling skills.\n- **Upstream-pinned.** Recipes live in\n  `https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills`, which evolves outside\n  this repo. Stale clones can drift; always `git pull` the upstream\n  before relying on a sibling skill.\n- **Gated content.** `nvidia\u002FPhysicalAI-*`, `nvidia\u002FDiffusionHarmonizer`,\n  and `nvidia\u002Fasset-harvester` require the user to accept license\n  terms on Hugging Face first. The router cannot bypass this.\n- **Heavy footprint.** A complete NuRec workflow can leave 150 GB+\n  on disk. See [`references\u002Fteardown.md`](references\u002Fteardown.md).\n- **NVIDIA-only stack.** Requires an NVIDIA GPU plus the NVIDIA\n  Container Toolkit. AMD \u002F Intel \u002F Apple Silicon are not supported.\n- **Not a SimReady pipeline.** NuRec produces a renderable USDZ from\n  a recording; SimReady packaging of CAD or source meshes is a\n  different pipeline (see `omniverse-cad-to-simready`).\n\n## Troubleshooting\n\n| Error \u002F symptom | Likely cause | Solution |\n|-----------------|--------------|----------|\n| `nurec-skills` clone missing or empty | Upstream not fetched yet | Run the clone block in [Locate and fetch the upstream skills](#locate-and-fetch-the-upstream-skills) |\n| `403`\u002F`401` pulling `nvidia\u002FPhysicalAI-*` from HF | Gated license not accepted, or `HF_TOKEN` unset \u002F wrong scope | Accept the gated license on Hugging Face, then `hf auth login` with a token that has `read` access |\n| `denied: requested access to the resource is denied` from `nvcr.io\u002Fnvidia\u002Fnre\u002F*` | Missing or expired `NGC_API_KEY` | `docker login nvcr.io` with `$oauthtoken` \u002F `NGC_API_KEY`; rotate the key at `org.ngc.nvidia.com\u002Fsetup\u002Fapi-key` if needed |\n| NRE refuses to load a clip (\"not valid NCore V4\") | Recording was not converted | Run the `ncore` skill before invoking `nre` |\n| `serve-grpc` cold-start latency dominates a Python loop | One-shot Docker invocation per render | Use the `nre` warm `serve-grpc` + thin Python client (`batch_render_rgb`) recipe |\n| Output files are owned by `root` after a `docker run` | `-u $(id -u):$(id -g)` was missing | `sudo chown -R \"$(id -u):$(id -g)\" \u003Coutput_dir>`; add the `-u` flag next time |\n| Frames have ghosting \u002F floaters \u002F flicker after rendering | Inline cleanup not enabled | Re-render with `nre --enable-difix`, or post-process with `nurec-fixer` (DiffusionHarmonizer) |\n| Stale skill names (`ncore-data-conversion`, old `nvidia\u002FFixer`) in agent output | Out-of-date cached skill | Update references to `ncore` and `nurec-fixer` (DiffusionHarmonizer); see [`references\u002Fmaintenance.md`](references\u002Fmaintenance.md) |\n| Bash anti-pattern `${HF_TOKEN:+yes}${HF_TOKEN:-no}` echoed token value | Misuse of bash parameter expansion | Rotate the token; use `hf auth whoami` or length-only checks (see [`references\u002Fsecrets-handling.md`](references\u002Fsecrets-handling.md)) |\n\n## Cross-skill teardown\n\nA complete NuRec workflow can leave **150 GB+** on disk between\ncontainer images, model weights, code clones, conda envs, and output\ndirectories. Each sibling skill has its own dedicated `Teardown`\nsection — read them in the order documented in\n[`references\u002Fteardown.md`](references\u002Fteardown.md) when the user no\nlonger needs the workflow.\n\n## Keeping this router up to date\n\nProcedure for adding new sibling skills, renames, or upstream URL\nchanges lives in [`references\u002Fmaintenance.md`](references\u002Fmaintenance.md).\nTreat the upstream `nurec-index` at\n\u003Chttps:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002FSKILL.md>\nas authoritative; this skill mirrors only the picker tables, the\nworkflow ordering, and the upstream fetch recipe.\n",{"data":35,"body":78},{"name":4,"description":6,"license":26,"version":36,"tools":37,"compatibility":40,"metadata":41},"0.3.0",[38,39],"Read","Shell","Router skill; downstream sibling skills require Docker, NVIDIA Container Toolkit, GPU, NGC API key, Hugging Face token with PhysicalAI gated licenses accepted, Python 3.10+, and `huggingface_hub`. Optional: CARLA \u002F Isaac Sim 5.1 \u002F AlpaSim for simulator integration over `serve-grpc`.",{"author":42,"tags":43,"upstream":47,"upstream_clone_path":76,"upstream_override_env":77},"NVIDIA Physical AI",[44,45,46],"physical-ai","nurec","neural-reconstruction",{"repo":48,"branch":49,"skills_dir":50,"skills_dir_alias":51,"index_skill":52,"index_skill_name":53,"sibling_skills":54},"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills","main",".agents\u002Fskills\u002F","skills\u002F",".agents\u002Fskills\u002FSKILL.md","nurec-index",[55,59,63,67,71],{"name":56,"folder":57,"upstream":58},"physical-ai-datasets","physical-ai-datasets\u002F","https:\u002F\u002Fhuggingface.co\u002Fnvidia",{"name":60,"folder":61,"upstream":62},"ncore","ncore\u002F","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fncore",{"name":64,"folder":65,"upstream":66},"nre","nre\u002F","nvcr.io\u002Fnvidia\u002Fnre\u002Fnre",{"name":68,"folder":69,"upstream":70},"asset-harvester","asset-harvester\u002F","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fasset-harvester",{"name":72,"folder":73,"upstream":74,"hf_model":75},"nurec-fixer","nurec-fixer\u002F","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fharmonizer","https:\u002F\u002Fhuggingface.co\u002Fnvidia\u002FDiffusionHarmonizer","${PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT:-$HOME\u002F.physical-ai-skill-hub\u002Fupstreams}\u002Fnurec-skills","NUREC_SKILLS_UPSTREAM_ROOT",{"type":79,"children":80},"root",[81,90,97,159,187,197,205,245,251,263,455,469,475,488,642,647,854,870,876,886,933,938,997,1010,1016,1021,1516,1522,1538,1601,1606,1612,1791,1806,1811,1827,2198,2203,2269,2274,2321,2327,2485,2491,2603,2609,3044,3050,3080,3086,3117],{"type":82,"tag":83,"props":84,"children":86},"element","h1",{"id":85},"physical-ai-neural-reconstruction-nurec-router",[87],{"type":88,"value":89},"text","Physical AI Neural Reconstruction (NuRec) Router",{"type":82,"tag":91,"props":92,"children":94},"h2",{"id":93},"purpose",[95],{"type":88,"value":96},"Purpose",{"type":82,"tag":98,"props":99,"children":100},"p",{},[101,103,109,111,117,119,124,126,131,133,138,139,144,145,150,152,157],{"type":88,"value":102},"This is a ",{"type":82,"tag":104,"props":105,"children":106},"strong",{},[107],{"type":88,"value":108},"thin router",{"type":88,"value":110}," for NVIDIA Neural Reconstruction (NuRec)\nrequests. It points at the upstream ",{"type":82,"tag":112,"props":113,"children":115},"code",{"className":114},[],[116],{"type":88,"value":53},{"type":88,"value":118}," skill at\n",{"type":82,"tag":112,"props":120,"children":122},{"className":121},[],[123],{"type":88,"value":48},{"type":88,"value":125}," and its five sibling skills\n(",{"type":82,"tag":112,"props":127,"children":129},{"className":128},[],[130],{"type":88,"value":56},{"type":88,"value":132},", ",{"type":82,"tag":112,"props":134,"children":136},{"className":135},[],[137],{"type":88,"value":60},{"type":88,"value":132},{"type":82,"tag":112,"props":140,"children":142},{"className":141},[],[143],{"type":88,"value":64},{"type":88,"value":132},{"type":82,"tag":112,"props":146,"children":148},{"className":147},[],[149],{"type":88,"value":68},{"type":88,"value":151},",\n",{"type":82,"tag":112,"props":153,"children":155},{"className":154},[],[156],{"type":88,"value":72},{"type":88,"value":158},"). Use this skill to:",{"type":82,"tag":160,"props":161,"children":162},"ul",{},[163,169,182],{"type":82,"tag":164,"props":165,"children":166},"li",{},[167],{"type":88,"value":168},"Identify which upstream sibling skill answers a NuRec question.",{"type":82,"tag":164,"props":170,"children":171},{},[172,174,180],{"type":88,"value":173},"Locate, clone, or refresh the canonical ",{"type":82,"tag":112,"props":175,"children":177},{"className":176},[],[178],{"type":88,"value":179},"nurec-skills",{"type":88,"value":181}," checkout.",{"type":82,"tag":164,"props":183,"children":184},{},[185],{"type":88,"value":186},"Order multi-step NuRec workflows (data → conversion → train →\nrender → cleanup) before opening the upstream recipe.",{"type":82,"tag":98,"props":188,"children":189},{},[190,192],{"type":88,"value":191},"The canonical recipes (training, rendering, data conversion, dataset\ndownloads, object harvesting, frame cleanup) live in the upstream\nsibling skills. ",{"type":82,"tag":104,"props":193,"children":194},{},[195],{"type":88,"value":196},"Never copy or reconstruct their commands here.",{"type":82,"tag":98,"props":198,"children":199},{},[200],{"type":82,"tag":104,"props":201,"children":202},{},[203],{"type":88,"value":204},"Do NOT use this skill for:",{"type":82,"tag":160,"props":206,"children":207},{},[208,221,233],{"type":82,"tag":164,"props":209,"children":210},{},[211,213,219],{"type":88,"value":212},"SimReady packaging of CAD or source meshes → use\n",{"type":82,"tag":112,"props":214,"children":216},{"className":215},[],[217],{"type":88,"value":218},"omniverse-cad-to-simready",{"type":88,"value":220},".",{"type":82,"tag":164,"props":222,"children":223},{},[224,226,232],{"type":88,"value":225},"Generic USD performance tuning unrelated to NuRec → use\n",{"type":82,"tag":112,"props":227,"children":229},{"className":228},[],[230],{"type":88,"value":231},"omniverse-usd-performance-tuning",{"type":88,"value":220},{"type":82,"tag":164,"props":234,"children":235},{},[236,238,244],{"type":88,"value":237},"AKS \u002F OSMO \u002F NIM Operator infrastructure setup → use\n",{"type":82,"tag":112,"props":239,"children":241},{"className":240},[],[242],{"type":88,"value":243},"physical-ai-infrastructure-setup-and-resilient-scaling",{"type":88,"value":220},{"type":82,"tag":91,"props":246,"children":248},{"id":247},"when-to-use",[249],{"type":88,"value":250},"When to Use",{"type":82,"tag":98,"props":252,"children":253},{},[254,256,261],{"type":88,"value":255},"Read this skill ",{"type":82,"tag":104,"props":257,"children":258},{},[259],{"type":88,"value":260},"first",{"type":88,"value":262}," whenever a user mentions any of:",{"type":82,"tag":98,"props":264,"children":265},{},[266,271,272,278,279,285,286,292,293,299,300,306,307,313,314,320,321,327,328,334,335,341,342,348,349,355,356,362,363,369,370,376,377,383,384,390,391,397,398,404,405,411,412,418,419,425,426,432,433,439,440,446,447,453],{"type":82,"tag":112,"props":267,"children":269},{"className":268},[],[270],{"type":88,"value":45},{"type":88,"value":132},{"type":82,"tag":112,"props":273,"children":275},{"className":274},[],[276],{"type":88,"value":277},"nurec router",{"type":88,"value":132},{"type":82,"tag":112,"props":280,"children":282},{"className":281},[],[283],{"type":88,"value":284},"nurec index",{"type":88,"value":132},{"type":82,"tag":112,"props":287,"children":289},{"className":288},[],[290],{"type":88,"value":291},"neural reconstruction",{"type":88,"value":151},{"type":82,"tag":112,"props":294,"children":296},{"className":295},[],[297],{"type":88,"value":298},"neural reconstruction engine",{"type":88,"value":132},{"type":82,"tag":112,"props":301,"children":303},{"className":302},[],[304],{"type":88,"value":305},"NRE",{"type":88,"value":132},{"type":82,"tag":112,"props":308,"children":310},{"className":309},[],[311],{"type":88,"value":312},"3DGUT",{"type":88,"value":132},{"type":82,"tag":112,"props":315,"children":317},{"className":316},[],[318],{"type":88,"value":319},"3DGRT",{"type":88,"value":132},{"type":82,"tag":112,"props":322,"children":324},{"className":323},[],[325],{"type":88,"value":326},"USDZ",{"type":88,"value":151},{"type":82,"tag":112,"props":329,"children":331},{"className":330},[],[332],{"type":88,"value":333},"NCore V4",{"type":88,"value":132},{"type":82,"tag":112,"props":336,"children":338},{"className":337},[],[339],{"type":88,"value":340},"sensor sim",{"type":88,"value":132},{"type":82,"tag":112,"props":343,"children":345},{"className":344},[],[346],{"type":88,"value":347},"novel view synthesis",{"type":88,"value":151},{"type":82,"tag":112,"props":350,"children":352},{"className":351},[],[353],{"type":88,"value":354},"PhysicalAI-Autonomous-Vehicles-NuRec",{"type":88,"value":132},{"type":82,"tag":112,"props":357,"children":359},{"className":358},[],[360],{"type":88,"value":361},"PhysicalAI-NuRec-PPISP",{"type":88,"value":151},{"type":82,"tag":112,"props":364,"children":366},{"className":365},[],[367],{"type":88,"value":368},"Cosmos-Drive-Dreams",{"type":88,"value":132},{"type":82,"tag":112,"props":371,"children":373},{"className":372},[],[374],{"type":88,"value":375},"asset harvester",{"type":88,"value":132},{"type":82,"tag":112,"props":378,"children":380},{"className":379},[],[381],{"type":88,"value":382},"nurec fixer",{"type":88,"value":151},{"type":82,"tag":112,"props":385,"children":387},{"className":386},[],[388],{"type":88,"value":389},"DiffusionHarmonizer",{"type":88,"value":132},{"type":82,"tag":112,"props":392,"children":394},{"className":393},[],[395],{"type":88,"value":396},"harmonizer",{"type":88,"value":132},{"type":82,"tag":112,"props":399,"children":401},{"className":400},[],[402],{"type":88,"value":403},"difix",{"type":88,"value":132},{"type":82,"tag":112,"props":406,"children":408},{"className":407},[],[409],{"type":88,"value":410},"difix3d",{"type":88,"value":132},{"type":82,"tag":112,"props":413,"children":415},{"className":414},[],[416],{"type":88,"value":417},"serve-grpc",{"type":88,"value":151},{"type":82,"tag":112,"props":420,"children":422},{"className":421},[],[423],{"type":88,"value":424},"render-grpc",{"type":88,"value":132},{"type":82,"tag":112,"props":427,"children":429},{"className":428},[],[430],{"type":88,"value":431},"warm serve-grpc",{"type":88,"value":132},{"type":82,"tag":112,"props":434,"children":436},{"className":435},[],[437],{"type":88,"value":438},"nre thin client",{"type":88,"value":132},{"type":82,"tag":112,"props":441,"children":443},{"className":442},[],[444],{"type":88,"value":445},"batch_render_rgb",{"type":88,"value":151},{"type":82,"tag":112,"props":448,"children":450},{"className":449},[],[451],{"type":88,"value":452},"nurec teardown",{"type":88,"value":454},", \"where do I start with NuRec\", \"which NuRec skill\nshould I use for X?\".",{"type":82,"tag":98,"props":456,"children":457},{},[458,460,467],{"type":88,"value":459},"Decide which upstream sibling skill answers the question, fetch it\n(see ",{"type":82,"tag":461,"props":462,"children":464},"a",{"href":463},"#locate-and-fetch-the-upstream-skills",[465],{"type":88,"value":466},"Locate and fetch the upstream skills",{"type":88,"value":468},"),\nthen follow that skill's body.",{"type":82,"tag":91,"props":470,"children":472},{"id":471},"prerequisites",[473],{"type":88,"value":474},"Prerequisites",{"type":82,"tag":98,"props":476,"children":477},{},[478,480,486],{"type":88,"value":479},"Router skill itself has no runtime prerequisites beyond ",{"type":82,"tag":112,"props":481,"children":483},{"className":482},[],[484],{"type":88,"value":485},"git",{"type":88,"value":487}," for\nfetching the upstream. Downstream sibling skills require:",{"type":82,"tag":160,"props":489,"children":490},{},[491,543,561,608,626],{"type":82,"tag":164,"props":492,"children":493},{},[494,499,501,506,507,513,515,520,522,527,528,534,535,541],{"type":82,"tag":104,"props":495,"children":496},{},[497],{"type":88,"value":498},"Docker + NVIDIA Container Toolkit + GPU",{"type":88,"value":500}," — for ",{"type":82,"tag":112,"props":502,"children":504},{"className":503},[],[505],{"type":88,"value":64},{"type":88,"value":132},{"type":82,"tag":112,"props":508,"children":510},{"className":509},[],[511],{"type":88,"value":512},"nre-tools",{"type":88,"value":514},",\nand ",{"type":82,"tag":112,"props":516,"children":518},{"className":517},[],[519],{"type":88,"value":72},{"type":88,"value":521}," containers\n(",{"type":82,"tag":112,"props":523,"children":525},{"className":524},[],[526],{"type":88,"value":66},{"type":88,"value":132},{"type":82,"tag":112,"props":529,"children":531},{"className":530},[],[532],{"type":88,"value":533},"nvcr.io\u002Fnvidia\u002Fnre\u002Fnre-tools",{"type":88,"value":151},{"type":82,"tag":112,"props":536,"children":538},{"className":537},[],[539],{"type":88,"value":540},"nvcr.io\u002Fnvidia\u002Fcosmos\u002Fcosmos-predict2-container:1.2",{"type":88,"value":542},").",{"type":82,"tag":164,"props":544,"children":545},{},[546,551,553,559],{"type":82,"tag":104,"props":547,"children":548},{},[549],{"type":88,"value":550},"NGC API key",{"type":88,"value":552}," (",{"type":82,"tag":112,"props":554,"children":556},{"className":555},[],[557],{"type":88,"value":558},"NGC_API_KEY",{"type":88,"value":560},") — for pulling NGC containers.",{"type":82,"tag":164,"props":562,"children":563},{},[564,569,570,576,578,584,585,591,593,599,601,606],{"type":82,"tag":104,"props":565,"children":566},{},[567],{"type":88,"value":568},"Hugging Face token",{"type":88,"value":552},{"type":82,"tag":112,"props":571,"children":573},{"className":572},[],[574],{"type":88,"value":575},"HF_TOKEN",{"type":88,"value":577},") with the\n",{"type":82,"tag":112,"props":579,"children":581},{"className":580},[],[582],{"type":88,"value":583},"nvidia\u002FPhysicalAI-*",{"type":88,"value":132},{"type":82,"tag":112,"props":586,"children":588},{"className":587},[],[589],{"type":88,"value":590},"nvidia\u002FDiffusionHarmonizer",{"type":88,"value":592},", and\n",{"type":82,"tag":112,"props":594,"children":596},{"className":595},[],[597],{"type":88,"value":598},"nvidia\u002Fasset-harvester",{"type":88,"value":600}," gated licenses ",{"type":82,"tag":104,"props":602,"children":603},{},[604],{"type":88,"value":605},"accepted in advance",{"type":88,"value":607}," on\nHugging Face.",{"type":82,"tag":164,"props":609,"children":610},{},[611,616,618,624],{"type":82,"tag":104,"props":612,"children":613},{},[614],{"type":88,"value":615},"Python 3.10+",{"type":88,"value":617}," with ",{"type":82,"tag":112,"props":619,"children":621},{"className":620},[],[622],{"type":88,"value":623},"huggingface_hub",{"type":88,"value":625}," installed.",{"type":82,"tag":164,"props":627,"children":628},{},[629,634,636,641],{"type":82,"tag":104,"props":630,"children":631},{},[632],{"type":88,"value":633},"(Optional)",{"type":88,"value":635}," CARLA, Isaac Sim 5.1, or AlpaSim for simulator\nintegration over ",{"type":82,"tag":112,"props":637,"children":639},{"className":638},[],[640],{"type":88,"value":417},{"type":88,"value":220},{"type":82,"tag":98,"props":643,"children":644},{},[645],{"type":88,"value":646},"Verify secrets safely (do not echo values):",{"type":82,"tag":648,"props":649,"children":654},"pre",{"className":650,"code":651,"language":652,"meta":653,"style":653},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","hf auth whoami\n[ -n \"${HF_TOKEN:-}\" ]      && echo \"HF_TOKEN length=${#HF_TOKEN}\"      || echo \"HF_TOKEN unset\"\n[ -n \"${NGC_API_KEY:-}\" ]   && echo \"NGC_API_KEY length=${#NGC_API_KEY}\" || echo \"NGC_API_KEY unset\"\n","bash","",[655],{"type":82,"tag":112,"props":656,"children":657},{"__ignoreMap":653},[658,681,774],{"type":82,"tag":659,"props":660,"children":663},"span",{"class":661,"line":662},"line",1,[664,670,676],{"type":82,"tag":659,"props":665,"children":667},{"style":666},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[668],{"type":88,"value":669},"hf",{"type":82,"tag":659,"props":671,"children":673},{"style":672},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[674],{"type":88,"value":675}," auth",{"type":82,"tag":659,"props":677,"children":678},{"style":672},[679],{"type":88,"value":680}," whoami\n",{"type":82,"tag":659,"props":682,"children":684},{"class":661,"line":683},2,[685,691,696,701,706,711,716,721,727,732,737,742,746,751,756,760,764,769],{"type":82,"tag":659,"props":686,"children":688},{"style":687},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[689],{"type":88,"value":690},"[",{"type":82,"tag":659,"props":692,"children":693},{"style":687},[694],{"type":88,"value":695}," -n",{"type":82,"tag":659,"props":697,"children":698},{"style":687},[699],{"type":88,"value":700}," \"${",{"type":82,"tag":659,"props":702,"children":704},{"style":703},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[705],{"type":88,"value":575},{"type":82,"tag":659,"props":707,"children":708},{"style":687},[709],{"type":88,"value":710},":-}\"",{"type":82,"tag":659,"props":712,"children":713},{"style":687},[714],{"type":88,"value":715}," ]",{"type":82,"tag":659,"props":717,"children":718},{"style":687},[719],{"type":88,"value":720},"      &&",{"type":82,"tag":659,"props":722,"children":724},{"style":723},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[725],{"type":88,"value":726}," echo",{"type":82,"tag":659,"props":728,"children":729},{"style":687},[730],{"type":88,"value":731}," \"",{"type":82,"tag":659,"props":733,"children":734},{"style":672},[735],{"type":88,"value":736},"HF_TOKEN length=",{"type":82,"tag":659,"props":738,"children":739},{"style":687},[740],{"type":88,"value":741},"${#",{"type":82,"tag":659,"props":743,"children":744},{"style":703},[745],{"type":88,"value":575},{"type":82,"tag":659,"props":747,"children":748},{"style":687},[749],{"type":88,"value":750},"}\"",{"type":82,"tag":659,"props":752,"children":753},{"style":687},[754],{"type":88,"value":755},"      ||",{"type":82,"tag":659,"props":757,"children":758},{"style":723},[759],{"type":88,"value":726},{"type":82,"tag":659,"props":761,"children":762},{"style":687},[763],{"type":88,"value":731},{"type":82,"tag":659,"props":765,"children":766},{"style":672},[767],{"type":88,"value":768},"HF_TOKEN unset",{"type":82,"tag":659,"props":770,"children":771},{"style":687},[772],{"type":88,"value":773},"\"\n",{"type":82,"tag":659,"props":775,"children":777},{"class":661,"line":776},3,[778,782,786,790,794,798,802,807,811,815,820,824,828,832,837,841,845,850],{"type":82,"tag":659,"props":779,"children":780},{"style":687},[781],{"type":88,"value":690},{"type":82,"tag":659,"props":783,"children":784},{"style":687},[785],{"type":88,"value":695},{"type":82,"tag":659,"props":787,"children":788},{"style":687},[789],{"type":88,"value":700},{"type":82,"tag":659,"props":791,"children":792},{"style":703},[793],{"type":88,"value":558},{"type":82,"tag":659,"props":795,"children":796},{"style":687},[797],{"type":88,"value":710},{"type":82,"tag":659,"props":799,"children":800},{"style":687},[801],{"type":88,"value":715},{"type":82,"tag":659,"props":803,"children":804},{"style":687},[805],{"type":88,"value":806},"   &&",{"type":82,"tag":659,"props":808,"children":809},{"style":723},[810],{"type":88,"value":726},{"type":82,"tag":659,"props":812,"children":813},{"style":687},[814],{"type":88,"value":731},{"type":82,"tag":659,"props":816,"children":817},{"style":672},[818],{"type":88,"value":819},"NGC_API_KEY length=",{"type":82,"tag":659,"props":821,"children":822},{"style":687},[823],{"type":88,"value":741},{"type":82,"tag":659,"props":825,"children":826},{"style":703},[827],{"type":88,"value":558},{"type":82,"tag":659,"props":829,"children":830},{"style":687},[831],{"type":88,"value":750},{"type":82,"tag":659,"props":833,"children":834},{"style":687},[835],{"type":88,"value":836}," ||",{"type":82,"tag":659,"props":838,"children":839},{"style":723},[840],{"type":88,"value":726},{"type":82,"tag":659,"props":842,"children":843},{"style":687},[844],{"type":88,"value":731},{"type":82,"tag":659,"props":846,"children":847},{"style":672},[848],{"type":88,"value":849},"NGC_API_KEY unset",{"type":82,"tag":659,"props":851,"children":852},{"style":687},[853],{"type":88,"value":773},{"type":82,"tag":98,"props":855,"children":856},{},[857,859,868],{"type":88,"value":858},"See ",{"type":82,"tag":461,"props":860,"children":862},{"href":861},"references\u002Fsecrets-handling.md",[863],{"type":82,"tag":112,"props":864,"children":866},{"className":865},[],[867],{"type":88,"value":861},{"type":88,"value":869},"\nfor the bash anti-patterns to avoid.",{"type":82,"tag":91,"props":871,"children":873},{"id":872},"what-is-nurec",[874],{"type":88,"value":875},"What is NuRec?",{"type":82,"tag":98,"props":877,"children":878},{},[879,884],{"type":82,"tag":104,"props":880,"children":881},{},[882],{"type":88,"value":883},"NuRec",{"type":88,"value":885}," (NVIDIA Omniverse Neural Reconstruction) takes camera, LiDAR,\nradar, or stereo recordings — typically from a self-driving car or a\nrobot — and turns them into a 3D scene you can re-render from any\nviewpoint. Names that come up a lot:",{"type":82,"tag":160,"props":887,"children":888},{},[889,905,914,923],{"type":82,"tag":164,"props":890,"children":891},{},[892,896,898,903],{"type":82,"tag":104,"props":893,"children":894},{},[895],{"type":88,"value":305},{"type":88,"value":897}," — \"Neural Reconstruction Engine\". NuRec is the product; NRE\nis the engine that trains and renders. Both route to the upstream\n",{"type":82,"tag":112,"props":899,"children":901},{"className":900},[],[902],{"type":88,"value":64},{"type":88,"value":904}," skill.",{"type":82,"tag":164,"props":906,"children":907},{},[908,912],{"type":82,"tag":104,"props":909,"children":910},{},[911],{"type":88,"value":326},{"type":88,"value":913}," — the file format of a trained scene. A zip archive that\nOmniverse, Isaac Sim, and CARLA can open.",{"type":82,"tag":164,"props":915,"children":916},{},[917,921],{"type":82,"tag":104,"props":918,"children":919},{},[920],{"type":88,"value":333},{"type":88,"value":922}," — the input format NRE consumes. Raw recordings must be\nconverted to NCore V4 before training.",{"type":82,"tag":164,"props":924,"children":925},{},[926,931],{"type":82,"tag":104,"props":927,"children":928},{},[929],{"type":88,"value":930},"3DGUT \u002F 3DGRT",{"type":88,"value":932}," — the two 3D Gaussian Splatting flavours used\ninternally by NRE. The default Hydra recipe picks one; most users\nnever set it manually.",{"type":82,"tag":98,"props":934,"children":935},{},[936],{"type":88,"value":937},"A typical NuRec project has three stages:",{"type":82,"tag":939,"props":940,"children":941},"ol",{},[942,965,981],{"type":82,"tag":164,"props":943,"children":944},{},[945,950,952,957,959,964],{"type":82,"tag":104,"props":946,"children":947},{},[948],{"type":88,"value":949},"Get the input",{"type":88,"value":951}," — convert your own recording to NCore V4\n(",{"type":82,"tag":112,"props":953,"children":955},{"className":954},[],[956],{"type":88,"value":60},{"type":88,"value":958},"), or download a pre-converted dataset\n(",{"type":82,"tag":112,"props":960,"children":962},{"className":961},[],[963],{"type":88,"value":56},{"type":88,"value":542},{"type":82,"tag":164,"props":966,"children":967},{},[968,973,975,980],{"type":82,"tag":104,"props":969,"children":970},{},[971],{"type":88,"value":972},"Train the reconstruction",{"type":88,"value":974}," — feed NCore V4 to NRE; out comes a\nUSDZ (",{"type":82,"tag":112,"props":976,"children":978},{"className":977},[],[979],{"type":88,"value":64},{"type":88,"value":542},{"type":82,"tag":164,"props":982,"children":983},{},[984,989,991,996],{"type":82,"tag":104,"props":985,"children":986},{},[987],{"type":88,"value":988},"Render new views",{"type":88,"value":990}," — render images, videos, or LiDAR sweeps from\nthe USDZ (",{"type":82,"tag":112,"props":992,"children":994},{"className":993},[],[995],{"type":88,"value":64},{"type":88,"value":542},{"type":82,"tag":98,"props":998,"children":999},{},[1000,1002,1008],{"type":88,"value":1001},"Projects that just want to ",{"type":82,"tag":1003,"props":1004,"children":1005},"em",{},[1006],{"type":88,"value":1007},"use",{"type":88,"value":1009}," an existing NVIDIA-published scene\nskip step 2.",{"type":82,"tag":91,"props":1011,"children":1013},{"id":1012},"pick-a-skill",[1014],{"type":88,"value":1015},"Pick a skill",{"type":82,"tag":98,"props":1017,"children":1018},{},[1019],{"type":88,"value":1020},"Match the user's goal in the left column and open the named upstream\nskill on the right. Arrows mean \"do these in order\".",{"type":82,"tag":1022,"props":1023,"children":1024},"table",{},[1025,1044],{"type":82,"tag":1026,"props":1027,"children":1028},"thead",{},[1029],{"type":82,"tag":1030,"props":1031,"children":1032},"tr",{},[1033,1039],{"type":82,"tag":1034,"props":1035,"children":1036},"th",{},[1037],{"type":88,"value":1038},"I want to…",{"type":82,"tag":1034,"props":1040,"children":1041},{},[1042],{"type":88,"value":1043},"Upstream skill",{"type":82,"tag":1045,"props":1046,"children":1047},"tbody",{},[1048,1065,1081,1097,1120,1145,1161,1179,1195,1219,1250,1274,1296,1312,1334,1373,1389,1413,1445,1469,1498],{"type":82,"tag":1030,"props":1049,"children":1050},{},[1051,1057],{"type":82,"tag":1052,"props":1053,"children":1054},"td",{},[1055],{"type":88,"value":1056},"Find or download a NuRec dataset NVIDIA has published",{"type":82,"tag":1052,"props":1058,"children":1059},{},[1060],{"type":82,"tag":112,"props":1061,"children":1063},{"className":1062},[],[1064],{"type":88,"value":56},{"type":82,"tag":1030,"props":1066,"children":1067},{},[1068,1073],{"type":82,"tag":1052,"props":1069,"children":1070},{},[1071],{"type":88,"value":1072},"Convert my own camera \u002F LiDAR \u002F radar \u002F depth \u002F stereo recording into NCore V4",{"type":82,"tag":1052,"props":1074,"children":1075},{},[1076],{"type":82,"tag":112,"props":1077,"children":1079},{"className":1078},[],[1080],{"type":88,"value":60},{"type":82,"tag":1030,"props":1082,"children":1083},{},[1084,1089],{"type":82,"tag":1052,"props":1085,"children":1086},{},[1087],{"type":88,"value":1088},"Write a new converter for an unsupported sensor setup (drone, RGB-D, ROS 2 bag, COLMAP, ScanNet++)",{"type":82,"tag":1052,"props":1090,"children":1091},{},[1092],{"type":82,"tag":112,"props":1093,"children":1095},{"className":1094},[],[1096],{"type":88,"value":60},{"type":82,"tag":1030,"props":1098,"children":1099},{},[1100,1105],{"type":82,"tag":1052,"props":1101,"children":1102},{},[1103],{"type":88,"value":1104},"Train a 3D reconstruction from an NCore clip",{"type":82,"tag":1052,"props":1106,"children":1107},{},[1108,1113,1115],{"type":82,"tag":112,"props":1109,"children":1111},{"className":1110},[],[1112],{"type":88,"value":60},{"type":88,"value":1114}," → ",{"type":82,"tag":112,"props":1116,"children":1118},{"className":1117},[],[1119],{"type":88,"value":64},{"type":82,"tag":1030,"props":1121,"children":1122},{},[1123,1128],{"type":82,"tag":1052,"props":1124,"children":1125},{},[1126],{"type":88,"value":1127},"Generate the extra inputs NRE needs (segmentation masks, depth, ego mask)",{"type":82,"tag":1052,"props":1129,"children":1130},{},[1131,1136,1138,1143],{"type":82,"tag":112,"props":1132,"children":1134},{"className":1133},[],[1135],{"type":88,"value":64},{"type":88,"value":1137}," (uses the ",{"type":82,"tag":112,"props":1139,"children":1141},{"className":1140},[],[1142],{"type":88,"value":512},{"type":88,"value":1144}," container)",{"type":82,"tag":1030,"props":1146,"children":1147},{},[1148,1153],{"type":82,"tag":1052,"props":1149,"children":1150},{},[1151],{"type":88,"value":1152},"Render a USDZ along the original camera positions",{"type":82,"tag":1052,"props":1154,"children":1155},{},[1156],{"type":82,"tag":112,"props":1157,"children":1159},{"className":1158},[],[1160],{"type":88,"value":64},{"type":82,"tag":1030,"props":1162,"children":1163},{},[1164,1169],{"type":82,"tag":1052,"props":1165,"children":1166},{},[1167],{"type":88,"value":1168},"Render at full resolution \u002F highest quality",{"type":82,"tag":1052,"props":1170,"children":1171},{},[1172,1177],{"type":82,"tag":112,"props":1173,"children":1175},{"className":1174},[],[1176],{"type":88,"value":64},{"type":88,"value":1178}," (see \"Quality presets\")",{"type":82,"tag":1030,"props":1180,"children":1181},{},[1182,1187],{"type":82,"tag":1052,"props":1183,"children":1184},{},[1185],{"type":88,"value":1186},"Render along a shifted trajectory (e.g. car moved 3 m left)",{"type":82,"tag":1052,"props":1188,"children":1189},{},[1190],{"type":82,"tag":112,"props":1191,"children":1193},{"className":1192},[],[1194],{"type":88,"value":64},{"type":82,"tag":1030,"props":1196,"children":1197},{},[1198,1203],{"type":82,"tag":1052,"props":1199,"children":1200},{},[1201],{"type":88,"value":1202},"Render through a server so CARLA \u002F Isaac Sim \u002F AlpaSim \u002F a custom simulator can ask for frames",{"type":82,"tag":1052,"props":1204,"children":1205},{},[1206,1211,1212,1217],{"type":82,"tag":112,"props":1207,"children":1209},{"className":1208},[],[1210],{"type":88,"value":64},{"type":88,"value":552},{"type":82,"tag":112,"props":1213,"children":1215},{"className":1214},[],[1216],{"type":88,"value":417},{"type":88,"value":1218},")",{"type":82,"tag":1030,"props":1220,"children":1221},{},[1222,1227],{"type":82,"tag":1052,"props":1223,"children":1224},{},[1225],{"type":88,"value":1226},"Render the same USDZ many times back-to-back from Python with minimal per-call latency",{"type":82,"tag":1052,"props":1228,"children":1229},{},[1230,1235,1237,1242,1244,1249],{"type":82,"tag":112,"props":1231,"children":1233},{"className":1232},[],[1234],{"type":88,"value":64},{"type":88,"value":1236}," (warm ",{"type":82,"tag":112,"props":1238,"children":1240},{"className":1239},[],[1241],{"type":88,"value":417},{"type":88,"value":1243}," + thin Python client \u002F ",{"type":82,"tag":112,"props":1245,"children":1247},{"className":1246},[],[1248],{"type":88,"value":445},{"type":88,"value":1218},{"type":82,"tag":1030,"props":1251,"children":1252},{},[1253,1258],{"type":82,"tag":1052,"props":1254,"children":1255},{},[1256],{"type":88,"value":1257},"Render LiDAR sweeps (point clouds) from a USDZ",{"type":82,"tag":1052,"props":1259,"children":1260},{},[1261,1266,1267,1273],{"type":82,"tag":112,"props":1262,"children":1264},{"className":1263},[],[1265],{"type":88,"value":64},{"type":88,"value":552},{"type":82,"tag":112,"props":1268,"children":1270},{"className":1269},[],[1271],{"type":88,"value":1272},"render-grpc --lidar",{"type":88,"value":1218},{"type":82,"tag":1030,"props":1275,"children":1276},{},[1277,1282],{"type":82,"tag":1052,"props":1278,"children":1279},{},[1280],{"type":88,"value":1281},"Skip training and just render a NuRec scene NVIDIA already built",{"type":82,"tag":1052,"props":1283,"children":1284},{},[1285,1290,1291],{"type":82,"tag":112,"props":1286,"children":1288},{"className":1287},[],[1289],{"type":88,"value":56},{"type":88,"value":1114},{"type":82,"tag":112,"props":1292,"children":1294},{"className":1293},[],[1295],{"type":88,"value":64},{"type":82,"tag":1030,"props":1297,"children":1298},{},[1299,1304],{"type":82,"tag":1052,"props":1300,"children":1301},{},[1302],{"type":88,"value":1303},"Extract individual 3D objects (cars, pedestrians) from a driving clip",{"type":82,"tag":1052,"props":1305,"children":1306},{},[1307],{"type":82,"tag":112,"props":1308,"children":1310},{"className":1309},[],[1311],{"type":88,"value":68},{"type":82,"tag":1030,"props":1313,"children":1314},{},[1315,1320],{"type":82,"tag":1052,"props":1316,"children":1317},{},[1318],{"type":88,"value":1319},"Add, remove, or replace cars \u002F pedestrians in a NuRec scene",{"type":82,"tag":1052,"props":1321,"children":1322},{},[1323,1328,1329],{"type":82,"tag":112,"props":1324,"children":1326},{"className":1325},[],[1327],{"type":88,"value":68},{"type":88,"value":1114},{"type":82,"tag":112,"props":1330,"children":1332},{"className":1331},[],[1333],{"type":88,"value":64},{"type":82,"tag":1030,"props":1335,"children":1336},{},[1337,1342],{"type":82,"tag":1052,"props":1338,"children":1339},{},[1340],{"type":88,"value":1341},"Clean up or harmonize rendered frames (ghosting, floaters, flicker, lighting\u002Fshadows)",{"type":82,"tag":1052,"props":1343,"children":1344},{},[1345,1350,1351,1356,1358,1364,1366,1371],{"type":82,"tag":112,"props":1346,"children":1348},{"className":1347},[],[1349],{"type":88,"value":72},{"type":88,"value":132},{"type":82,"tag":104,"props":1352,"children":1353},{},[1354],{"type":88,"value":1355},"or",{"type":88,"value":1357}," ",{"type":82,"tag":112,"props":1359,"children":1361},{"className":1360},[],[1362],{"type":88,"value":1363},"--enable-difix",{"type":88,"value":1365}," inside ",{"type":82,"tag":112,"props":1367,"children":1369},{"className":1368},[],[1370],{"type":88,"value":64},{"type":88,"value":1372}," for inline rendering",{"type":82,"tag":1030,"props":1374,"children":1375},{},[1376,1381],{"type":82,"tag":1052,"props":1377,"children":1378},{},[1379],{"type":88,"value":1380},"Export the scene as a PLY, mesh, depth maps, ego mask, etc.",{"type":82,"tag":1052,"props":1382,"children":1383},{},[1384],{"type":82,"tag":112,"props":1385,"children":1387},{"className":1386},[],[1388],{"type":88,"value":64},{"type":82,"tag":1030,"props":1390,"children":1391},{},[1392,1397],{"type":82,"tag":1052,"props":1393,"children":1394},{},[1395],{"type":88,"value":1396},"Upgrade an old USDZ so newer NRE versions load it faster",{"type":82,"tag":1052,"props":1398,"children":1399},{},[1400,1405,1406,1412],{"type":82,"tag":112,"props":1401,"children":1403},{"className":1402},[],[1404],{"type":88,"value":64},{"type":88,"value":552},{"type":82,"tag":112,"props":1407,"children":1409},{"className":1408},[],[1410],{"type":88,"value":1411},"upgrade-artifact",{"type":88,"value":1218},{"type":82,"tag":1030,"props":1414,"children":1415},{},[1416,1421],{"type":82,"tag":1052,"props":1417,"children":1418},{},[1419],{"type":88,"value":1420},"Open a USDZ or PLY in a browser viewer",{"type":82,"tag":1052,"props":1422,"children":1423},{},[1424,1429,1430,1436,1438,1444],{"type":82,"tag":112,"props":1425,"children":1427},{"className":1426},[],[1428],{"type":88,"value":64},{"type":88,"value":552},{"type":82,"tag":112,"props":1431,"children":1433},{"className":1432},[],[1434],{"type":88,"value":1435},"viewer",{"type":88,"value":1437}," \u002F ",{"type":82,"tag":112,"props":1439,"children":1441},{"className":1440},[],[1442],{"type":88,"value":1443},"ply_viewer",{"type":88,"value":1218},{"type":82,"tag":1030,"props":1446,"children":1447},{},[1448,1453],{"type":82,"tag":1052,"props":1449,"children":1450},{},[1451],{"type":88,"value":1452},"Measure rendering quality (PSNR, SSIM, LPIPS) against ground truth",{"type":82,"tag":1052,"props":1454,"children":1455},{},[1456,1461,1462,1468],{"type":82,"tag":112,"props":1457,"children":1459},{"className":1458},[],[1460],{"type":88,"value":64},{"type":88,"value":552},{"type":82,"tag":112,"props":1463,"children":1465},{"className":1464},[],[1466],{"type":88,"value":1467},"eval-rendering-metrics",{"type":88,"value":1218},{"type":82,"tag":1030,"props":1470,"children":1471},{},[1472,1477],{"type":82,"tag":1052,"props":1473,"children":1474},{},[1475],{"type":88,"value":1476},"Benchmark different reconstruction methods on the same scenes",{"type":82,"tag":1052,"props":1478,"children":1479},{},[1480,1485,1486,1491,1493],{"type":82,"tag":112,"props":1481,"children":1483},{"className":1482},[],[1484],{"type":88,"value":56},{"type":88,"value":552},{"type":82,"tag":112,"props":1487,"children":1489},{"className":1488},[],[1490],{"type":88,"value":361},{"type":88,"value":1492},") → ",{"type":82,"tag":112,"props":1494,"children":1496},{"className":1495},[],[1497],{"type":88,"value":64},{"type":82,"tag":1030,"props":1499,"children":1500},{},[1501,1506],{"type":82,"tag":1052,"props":1502,"children":1503},{},[1504],{"type":88,"value":1505},"Train on multiple GPUs or on SLURM",{"type":82,"tag":1052,"props":1507,"children":1508},{},[1509,1514],{"type":82,"tag":112,"props":1510,"children":1512},{"className":1511},[],[1513],{"type":88,"value":64},{"type":88,"value":1515}," (Workflow D)",{"type":82,"tag":91,"props":1517,"children":1519},{"id":1518},"common-workflows",[1520],{"type":88,"value":1521},"Common workflows",{"type":82,"tag":98,"props":1523,"children":1524},{},[1525,1527,1536],{"type":88,"value":1526},"Six end-to-end workflows are documented in\n",{"type":82,"tag":461,"props":1528,"children":1530},{"href":1529},"references\u002Fworkflows.md",[1531],{"type":82,"tag":112,"props":1532,"children":1534},{"className":1533},[],[1535],{"type":88,"value":1529},{"type":88,"value":1537},":",{"type":82,"tag":160,"props":1539,"children":1540},{},[1541,1551,1561,1571,1581,1591],{"type":82,"tag":164,"props":1542,"children":1543},{},[1544,1549],{"type":82,"tag":104,"props":1545,"children":1546},{},[1547],{"type":88,"value":1548},"A.",{"type":88,"value":1550}," Make a NuRec scene from your own recording.",{"type":82,"tag":164,"props":1552,"children":1553},{},[1554,1559],{"type":82,"tag":104,"props":1555,"children":1556},{},[1557],{"type":88,"value":1558},"B.",{"type":88,"value":1560}," Use a NuRec scene NVIDIA has already trained.",{"type":82,"tag":164,"props":1562,"children":1563},{},[1564,1569],{"type":82,"tag":104,"props":1565,"children":1566},{},[1567],{"type":88,"value":1568},"C.",{"type":88,"value":1570}," Add, remove, or replace 3D objects in a scene.",{"type":82,"tag":164,"props":1572,"children":1573},{},[1574,1579],{"type":82,"tag":104,"props":1575,"children":1576},{},[1577],{"type":88,"value":1578},"D.",{"type":88,"value":1580}," Clean up rendered frames.",{"type":82,"tag":164,"props":1582,"children":1583},{},[1584,1589],{"type":82,"tag":104,"props":1585,"children":1586},{},[1587],{"type":88,"value":1588},"E.",{"type":88,"value":1590}," Benchmark reconstruction quality.",{"type":82,"tag":164,"props":1592,"children":1593},{},[1594,1599],{"type":82,"tag":104,"props":1595,"children":1596},{},[1597],{"type":88,"value":1598},"F.",{"type":88,"value":1600}," Connect NuRec to a simulator.",{"type":82,"tag":98,"props":1602,"children":1603},{},[1604],{"type":88,"value":1605},"Open that file when the user's task spans more than one sibling skill.",{"type":82,"tag":91,"props":1607,"children":1609},{"id":1608},"sibling-skills-upstream",[1610],{"type":88,"value":1611},"Sibling skills (upstream)",{"type":82,"tag":1022,"props":1613,"children":1614},{},[1615,1636],{"type":82,"tag":1026,"props":1616,"children":1617},{},[1618],{"type":82,"tag":1030,"props":1619,"children":1620},{},[1621,1626,1631],{"type":82,"tag":1034,"props":1622,"children":1623},{},[1624],{"type":88,"value":1625},"Name",{"type":82,"tag":1034,"props":1627,"children":1628},{},[1629],{"type":88,"value":1630},"Upstream folder",{"type":82,"tag":1034,"props":1632,"children":1633},{},[1634],{"type":88,"value":1635},"What it does",{"type":82,"tag":1045,"props":1637,"children":1638},{},[1639,1664,1689,1727,1760],{"type":82,"tag":1030,"props":1640,"children":1641},{},[1642,1650,1659],{"type":82,"tag":1052,"props":1643,"children":1644},{},[1645],{"type":82,"tag":112,"props":1646,"children":1648},{"className":1647},[],[1649],{"type":88,"value":56},{"type":82,"tag":1052,"props":1651,"children":1652},{},[1653],{"type":82,"tag":112,"props":1654,"children":1656},{"className":1655},[],[1657],{"type":88,"value":1658},".agents\u002Fskills\u002Fphysical-ai-datasets\u002F",{"type":82,"tag":1052,"props":1660,"children":1661},{},[1662],{"type":88,"value":1663},"Catalog and download recipes for every NVIDIA Physical AI dataset on Hugging Face (driving, robotics, manipulation, NuRec scenes, benchmarks).",{"type":82,"tag":1030,"props":1665,"children":1666},{},[1667,1675,1684],{"type":82,"tag":1052,"props":1668,"children":1669},{},[1670],{"type":82,"tag":112,"props":1671,"children":1673},{"className":1672},[],[1674],{"type":88,"value":60},{"type":82,"tag":1052,"props":1676,"children":1677},{},[1678],{"type":82,"tag":112,"props":1679,"children":1681},{"className":1680},[],[1682],{"type":88,"value":1683},".agents\u002Fskills\u002Fncore\u002F",{"type":82,"tag":1052,"props":1685,"children":1686},{},[1687],{"type":88,"value":1688},"Converts any sensor recording to NCore V4 (the format NRE needs). Also covers writing a new converter.",{"type":82,"tag":1030,"props":1690,"children":1691},{},[1692,1700,1709],{"type":82,"tag":1052,"props":1693,"children":1694},{},[1695],{"type":82,"tag":112,"props":1696,"children":1698},{"className":1697},[],[1699],{"type":88,"value":64},{"type":82,"tag":1052,"props":1701,"children":1702},{},[1703],{"type":82,"tag":112,"props":1704,"children":1706},{"className":1705},[],[1707],{"type":88,"value":1708},".agents\u002Fskills\u002Fnre\u002F",{"type":82,"tag":1052,"props":1710,"children":1711},{},[1712,1714,1719,1720,1725],{"type":88,"value":1713},"The Neural Reconstruction Engine itself. Trains, renders (locally, via warm ",{"type":82,"tag":112,"props":1715,"children":1717},{"className":1716},[],[1718],{"type":88,"value":417},{"type":88,"value":1243},{"type":82,"tag":112,"props":1721,"children":1723},{"className":1722},[],[1724],{"type":88,"value":445},{"type":88,"value":1726},", or to an external simulator), exports meshes \u002F point clouds \u002F depth, edits actors, evaluates quality.",{"type":82,"tag":1030,"props":1728,"children":1729},{},[1730,1738,1747],{"type":82,"tag":1052,"props":1731,"children":1732},{},[1733],{"type":82,"tag":112,"props":1734,"children":1736},{"className":1735},[],[1737],{"type":88,"value":68},{"type":82,"tag":1052,"props":1739,"children":1740},{},[1741],{"type":82,"tag":112,"props":1742,"children":1744},{"className":1743},[],[1745],{"type":88,"value":1746},".agents\u002Fskills\u002Fasset-harvester\u002F",{"type":82,"tag":1052,"props":1748,"children":1749},{},[1750,1752,1758],{"type":88,"value":1751},"Open-source Apache-2.0 pipeline that extracts individual 3D objects from sparse views in a driving clip and saves them as ",{"type":82,"tag":112,"props":1753,"children":1755},{"className":1754},[],[1756],{"type":88,"value":1757},".ply",{"type":88,"value":1759}," Gaussian splats with metadata.",{"type":82,"tag":1030,"props":1761,"children":1762},{},[1763,1771,1780],{"type":82,"tag":1052,"props":1764,"children":1765},{},[1766],{"type":82,"tag":112,"props":1767,"children":1769},{"className":1768},[],[1770],{"type":88,"value":72},{"type":82,"tag":1052,"props":1772,"children":1773},{},[1774],{"type":82,"tag":112,"props":1775,"children":1777},{"className":1776},[],[1778],{"type":88,"value":1779},".agents\u002Fskills\u002Fnurec-fixer\u002F",{"type":82,"tag":1052,"props":1781,"children":1782},{},[1783,1785,1789],{"type":88,"value":1784},"Standalone NVIDIA ",{"type":82,"tag":104,"props":1786,"children":1787},{},[1788],{"type":88,"value":389},{"type":88,"value":1790}," workflow — public successor to the older Fixer \u002F Difix3D+ recipes — that cleans rendered frames, harmonizes inserted actors, evaluates PSNR\u002FLPIPS, and optionally fine-tunes the model.",{"type":82,"tag":98,"props":1792,"children":1793},{},[1794,1796,1805],{"type":88,"value":1795},"For naming overlaps (NRE vs Fixer, ncore vs nre, AV-NuRec vs\nCosmos-Drive-Dreams, NuRec vs SimReady) see\n",{"type":82,"tag":461,"props":1797,"children":1799},{"href":1798},"references\u002Fmix-ups.md",[1800],{"type":82,"tag":112,"props":1801,"children":1803},{"className":1802},[],[1804],{"type":88,"value":1798},{"type":88,"value":220},{"type":82,"tag":91,"props":1807,"children":1809},{"id":1808},"locate-and-fetch-the-upstream-skills",[1810],{"type":88,"value":466},{"type":82,"tag":98,"props":1812,"children":1813},{},[1814,1816,1825],{"type":88,"value":1815},"Quick recipe (full version in\n",{"type":82,"tag":461,"props":1817,"children":1819},{"href":1818},"references\u002Fupstream-fetch.md",[1820],{"type":82,"tag":112,"props":1821,"children":1823},{"className":1822},[],[1824],{"type":88,"value":1818},{"type":88,"value":1826},"):",{"type":82,"tag":648,"props":1828,"children":1830},{"className":650,"code":1829,"language":652,"meta":653,"style":653},"UPSTREAM_ROOT=\"${NUREC_SKILLS_UPSTREAM_ROOT:-${PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT:-$HOME\u002F.physical-ai-skill-hub\u002Fupstreams}}\"\nmkdir -p \"$UPSTREAM_ROOT\"\nif [ -d \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.git\" ]; then\n  git -C \"$UPSTREAM_ROOT\u002Fnurec-skills\" fetch --tags\n  git -C \"$UPSTREAM_ROOT\u002Fnurec-skills\" checkout main\n  git -C \"$UPSTREAM_ROOT\u002Fnurec-skills\" pull --ff-only\nelse\n  git clone --depth 1 https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills.git \\\n    \"$UPSTREAM_ROOT\u002Fnurec-skills\"\nfi\ntest -f \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.agents\u002Fskills\u002FSKILL.md\"\n",[1831],{"type":82,"tag":112,"props":1832,"children":1833},{"__ignoreMap":653},[1834,1904,1930,1977,2018,2056,2094,2103,2137,2158,2167],{"type":82,"tag":659,"props":1835,"children":1836},{"class":661,"line":662},[1837,1842,1847,1852,1856,1861,1866,1871,1876,1881,1885,1890,1894,1899],{"type":82,"tag":659,"props":1838,"children":1839},{"style":703},[1840],{"type":88,"value":1841},"UPSTREAM_ROOT",{"type":82,"tag":659,"props":1843,"children":1844},{"style":687},[1845],{"type":88,"value":1846},"=",{"type":82,"tag":659,"props":1848,"children":1849},{"style":687},[1850],{"type":88,"value":1851},"\"${",{"type":82,"tag":659,"props":1853,"children":1854},{"style":703},[1855],{"type":88,"value":77},{"type":82,"tag":659,"props":1857,"children":1858},{"style":687},[1859],{"type":88,"value":1860},":-${",{"type":82,"tag":659,"props":1862,"children":1863},{"style":703},[1864],{"type":88,"value":1865},"PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT",{"type":82,"tag":659,"props":1867,"children":1868},{"style":687},[1869],{"type":88,"value":1870},":-",{"type":82,"tag":659,"props":1872,"children":1873},{"style":703},[1874],{"type":88,"value":1875},"$HOME",{"type":82,"tag":659,"props":1877,"children":1878},{"style":687},[1879],{"type":88,"value":1880},"\u002F",{"type":82,"tag":659,"props":1882,"children":1883},{"style":672},[1884],{"type":88,"value":220},{"type":82,"tag":659,"props":1886,"children":1887},{"style":703},[1888],{"type":88,"value":1889},"physical-ai-skill-hub",{"type":82,"tag":659,"props":1891,"children":1892},{"style":687},[1893],{"type":88,"value":1880},{"type":82,"tag":659,"props":1895,"children":1896},{"style":703},[1897],{"type":88,"value":1898},"upstreams",{"type":82,"tag":659,"props":1900,"children":1901},{"style":687},[1902],{"type":88,"value":1903},"}}\"\n",{"type":82,"tag":659,"props":1905,"children":1906},{"class":661,"line":683},[1907,1912,1917,1921,1926],{"type":82,"tag":659,"props":1908,"children":1909},{"style":666},[1910],{"type":88,"value":1911},"mkdir",{"type":82,"tag":659,"props":1913,"children":1914},{"style":672},[1915],{"type":88,"value":1916}," -p",{"type":82,"tag":659,"props":1918,"children":1919},{"style":687},[1920],{"type":88,"value":731},{"type":82,"tag":659,"props":1922,"children":1923},{"style":703},[1924],{"type":88,"value":1925},"$UPSTREAM_ROOT",{"type":82,"tag":659,"props":1927,"children":1928},{"style":687},[1929],{"type":88,"value":773},{"type":82,"tag":659,"props":1931,"children":1932},{"class":661,"line":776},[1933,1939,1944,1949,1953,1957,1962,1967,1972],{"type":82,"tag":659,"props":1934,"children":1936},{"style":1935},"--shiki-light:#39ADB5;--shiki-light-font-style:italic;--shiki-default:#89DDFF;--shiki-default-font-style:italic;--shiki-dark:#89DDFF;--shiki-dark-font-style:italic",[1937],{"type":88,"value":1938},"if",{"type":82,"tag":659,"props":1940,"children":1941},{"style":687},[1942],{"type":88,"value":1943}," [",{"type":82,"tag":659,"props":1945,"children":1946},{"style":687},[1947],{"type":88,"value":1948}," -d",{"type":82,"tag":659,"props":1950,"children":1951},{"style":687},[1952],{"type":88,"value":731},{"type":82,"tag":659,"props":1954,"children":1955},{"style":703},[1956],{"type":88,"value":1925},{"type":82,"tag":659,"props":1958,"children":1959},{"style":672},[1960],{"type":88,"value":1961},"\u002Fnurec-skills\u002F.git",{"type":82,"tag":659,"props":1963,"children":1964},{"style":687},[1965],{"type":88,"value":1966},"\"",{"type":82,"tag":659,"props":1968,"children":1969},{"style":687},[1970],{"type":88,"value":1971}," ];",{"type":82,"tag":659,"props":1973,"children":1974},{"style":1935},[1975],{"type":88,"value":1976}," then\n",{"type":82,"tag":659,"props":1978,"children":1980},{"class":661,"line":1979},4,[1981,1986,1991,1995,1999,2004,2008,2013],{"type":82,"tag":659,"props":1982,"children":1983},{"style":666},[1984],{"type":88,"value":1985},"  git",{"type":82,"tag":659,"props":1987,"children":1988},{"style":672},[1989],{"type":88,"value":1990}," -C",{"type":82,"tag":659,"props":1992,"children":1993},{"style":687},[1994],{"type":88,"value":731},{"type":82,"tag":659,"props":1996,"children":1997},{"style":703},[1998],{"type":88,"value":1925},{"type":82,"tag":659,"props":2000,"children":2001},{"style":672},[2002],{"type":88,"value":2003},"\u002Fnurec-skills",{"type":82,"tag":659,"props":2005,"children":2006},{"style":687},[2007],{"type":88,"value":1966},{"type":82,"tag":659,"props":2009,"children":2010},{"style":672},[2011],{"type":88,"value":2012}," fetch",{"type":82,"tag":659,"props":2014,"children":2015},{"style":672},[2016],{"type":88,"value":2017}," --tags\n",{"type":82,"tag":659,"props":2019,"children":2021},{"class":661,"line":2020},5,[2022,2026,2030,2034,2038,2042,2046,2051],{"type":82,"tag":659,"props":2023,"children":2024},{"style":666},[2025],{"type":88,"value":1985},{"type":82,"tag":659,"props":2027,"children":2028},{"style":672},[2029],{"type":88,"value":1990},{"type":82,"tag":659,"props":2031,"children":2032},{"style":687},[2033],{"type":88,"value":731},{"type":82,"tag":659,"props":2035,"children":2036},{"style":703},[2037],{"type":88,"value":1925},{"type":82,"tag":659,"props":2039,"children":2040},{"style":672},[2041],{"type":88,"value":2003},{"type":82,"tag":659,"props":2043,"children":2044},{"style":687},[2045],{"type":88,"value":1966},{"type":82,"tag":659,"props":2047,"children":2048},{"style":672},[2049],{"type":88,"value":2050}," checkout",{"type":82,"tag":659,"props":2052,"children":2053},{"style":672},[2054],{"type":88,"value":2055}," main\n",{"type":82,"tag":659,"props":2057,"children":2059},{"class":661,"line":2058},6,[2060,2064,2068,2072,2076,2080,2084,2089],{"type":82,"tag":659,"props":2061,"children":2062},{"style":666},[2063],{"type":88,"value":1985},{"type":82,"tag":659,"props":2065,"children":2066},{"style":672},[2067],{"type":88,"value":1990},{"type":82,"tag":659,"props":2069,"children":2070},{"style":687},[2071],{"type":88,"value":731},{"type":82,"tag":659,"props":2073,"children":2074},{"style":703},[2075],{"type":88,"value":1925},{"type":82,"tag":659,"props":2077,"children":2078},{"style":672},[2079],{"type":88,"value":2003},{"type":82,"tag":659,"props":2081,"children":2082},{"style":687},[2083],{"type":88,"value":1966},{"type":82,"tag":659,"props":2085,"children":2086},{"style":672},[2087],{"type":88,"value":2088}," pull",{"type":82,"tag":659,"props":2090,"children":2091},{"style":672},[2092],{"type":88,"value":2093}," --ff-only\n",{"type":82,"tag":659,"props":2095,"children":2097},{"class":661,"line":2096},7,[2098],{"type":82,"tag":659,"props":2099,"children":2100},{"style":1935},[2101],{"type":88,"value":2102},"else\n",{"type":82,"tag":659,"props":2104,"children":2106},{"class":661,"line":2105},8,[2107,2111,2116,2121,2127,2132],{"type":82,"tag":659,"props":2108,"children":2109},{"style":666},[2110],{"type":88,"value":1985},{"type":82,"tag":659,"props":2112,"children":2113},{"style":672},[2114],{"type":88,"value":2115}," clone",{"type":82,"tag":659,"props":2117,"children":2118},{"style":672},[2119],{"type":88,"value":2120}," --depth",{"type":82,"tag":659,"props":2122,"children":2124},{"style":2123},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[2125],{"type":88,"value":2126}," 1",{"type":82,"tag":659,"props":2128,"children":2129},{"style":672},[2130],{"type":88,"value":2131}," https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills.git",{"type":82,"tag":659,"props":2133,"children":2134},{"style":703},[2135],{"type":88,"value":2136}," \\\n",{"type":82,"tag":659,"props":2138,"children":2140},{"class":661,"line":2139},9,[2141,2146,2150,2154],{"type":82,"tag":659,"props":2142,"children":2143},{"style":687},[2144],{"type":88,"value":2145},"    \"",{"type":82,"tag":659,"props":2147,"children":2148},{"style":703},[2149],{"type":88,"value":1925},{"type":82,"tag":659,"props":2151,"children":2152},{"style":672},[2153],{"type":88,"value":2003},{"type":82,"tag":659,"props":2155,"children":2156},{"style":687},[2157],{"type":88,"value":773},{"type":82,"tag":659,"props":2159,"children":2161},{"class":661,"line":2160},10,[2162],{"type":82,"tag":659,"props":2163,"children":2164},{"style":1935},[2165],{"type":88,"value":2166},"fi\n",{"type":82,"tag":659,"props":2168,"children":2170},{"class":661,"line":2169},11,[2171,2176,2181,2185,2189,2194],{"type":82,"tag":659,"props":2172,"children":2173},{"style":723},[2174],{"type":88,"value":2175},"test",{"type":82,"tag":659,"props":2177,"children":2178},{"style":672},[2179],{"type":88,"value":2180}," -f",{"type":82,"tag":659,"props":2182,"children":2183},{"style":687},[2184],{"type":88,"value":731},{"type":82,"tag":659,"props":2186,"children":2187},{"style":703},[2188],{"type":88,"value":1925},{"type":82,"tag":659,"props":2190,"children":2191},{"style":672},[2192],{"type":88,"value":2193},"\u002Fnurec-skills\u002F.agents\u002Fskills\u002FSKILL.md",{"type":82,"tag":659,"props":2195,"children":2196},{"style":687},[2197],{"type":88,"value":773},{"type":82,"tag":98,"props":2199,"children":2200},{},[2201],{"type":88,"value":2202},"Then read the upstream skill before running any mutating command:",{"type":82,"tag":648,"props":2204,"children":2206},{"className":650,"code":2205,"language":652,"meta":653,"style":653},"cat \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.agents\u002Fskills\u002FSKILL.md\"          # router\ncat \"$UPSTREAM_ROOT\u002Fnurec-skills\u002F.agents\u002Fskills\u002F\u003Cfolder>\u002FSKILL.md\" # sibling\n",[2207],{"type":82,"tag":112,"props":2208,"children":2209},{"__ignoreMap":653},[2210,2240],{"type":82,"tag":659,"props":2211,"children":2212},{"class":661,"line":662},[2213,2218,2222,2226,2230,2234],{"type":82,"tag":659,"props":2214,"children":2215},{"style":666},[2216],{"type":88,"value":2217},"cat",{"type":82,"tag":659,"props":2219,"children":2220},{"style":687},[2221],{"type":88,"value":731},{"type":82,"tag":659,"props":2223,"children":2224},{"style":703},[2225],{"type":88,"value":1925},{"type":82,"tag":659,"props":2227,"children":2228},{"style":672},[2229],{"type":88,"value":2193},{"type":82,"tag":659,"props":2231,"children":2232},{"style":687},[2233],{"type":88,"value":1966},{"type":82,"tag":659,"props":2235,"children":2237},{"style":2236},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[2238],{"type":88,"value":2239},"          # router\n",{"type":82,"tag":659,"props":2241,"children":2242},{"class":661,"line":683},[2243,2247,2251,2255,2260,2264],{"type":82,"tag":659,"props":2244,"children":2245},{"style":666},[2246],{"type":88,"value":2217},{"type":82,"tag":659,"props":2248,"children":2249},{"style":687},[2250],{"type":88,"value":731},{"type":82,"tag":659,"props":2252,"children":2253},{"style":703},[2254],{"type":88,"value":1925},{"type":82,"tag":659,"props":2256,"children":2257},{"style":672},[2258],{"type":88,"value":2259},"\u002Fnurec-skills\u002F.agents\u002Fskills\u002F\u003Cfolder>\u002FSKILL.md",{"type":82,"tag":659,"props":2261,"children":2262},{"style":687},[2263],{"type":88,"value":1966},{"type":82,"tag":659,"props":2265,"children":2266},{"style":2236},[2267],{"type":88,"value":2268}," # sibling\n",{"type":82,"tag":98,"props":2270,"children":2271},{},[2272],{"type":88,"value":2273},"Local lookup order (try in order before the upstream clone):",{"type":82,"tag":939,"props":2275,"children":2276},{},[2277,2288,2299,2310],{"type":82,"tag":164,"props":2278,"children":2279},{},[2280,2286],{"type":82,"tag":112,"props":2281,"children":2283},{"className":2282},[],[2284],{"type":88,"value":2285},".agents\u002Fskills\u002F\u003Cname>\u002FSKILL.md",{"type":88,"value":2287}," (Cursor, Codex, NemoClaw)",{"type":82,"tag":164,"props":2289,"children":2290},{},[2291,2297],{"type":82,"tag":112,"props":2292,"children":2294},{"className":2293},[],[2295],{"type":88,"value":2296},".claude\u002Fskills\u002F\u003Cname>\u002FSKILL.md",{"type":88,"value":2298}," (Claude Code)",{"type":82,"tag":164,"props":2300,"children":2301},{},[2302,2308],{"type":82,"tag":112,"props":2303,"children":2305},{"className":2304},[],[2306],{"type":88,"value":2307},".cursor\u002Fskills\u002F\u003Cname>\u002FSKILL.md",{"type":88,"value":2309}," (project-scoped)",{"type":82,"tag":164,"props":2311,"children":2312},{},[2313,2319],{"type":82,"tag":112,"props":2314,"children":2316},{"className":2315},[],[2317],{"type":88,"value":2318},"~\u002F.cursor\u002Fskills\u002F\u003Cname>\u002FSKILL.md",{"type":88,"value":2320}," (personal skills)",{"type":82,"tag":91,"props":2322,"children":2324},{"id":2323},"hard-rules",[2325],{"type":88,"value":2326},"Hard Rules",{"type":82,"tag":160,"props":2328,"children":2329},{},[2330,2335,2355,2383,2401,2435,2461,2474],{"type":82,"tag":164,"props":2331,"children":2332},{},[2333],{"type":88,"value":2334},"Router only — do not duplicate upstream NuRec recipes here. Read\nthe upstream sibling skill body before running any mutating command.",{"type":82,"tag":164,"props":2336,"children":2337},{},[2338,2340,2346,2348,2353],{"type":88,"value":2339},"Refer to sibling skills by their ",{"type":82,"tag":112,"props":2341,"children":2343},{"className":2342},[],[2344],{"type":88,"value":2345},"name:",{"type":88,"value":2347}," (e.g. ",{"type":82,"tag":112,"props":2349,"children":2351},{"className":2350},[],[2352],{"type":88,"value":64},{"type":88,"value":2354},"), not by repo\npath. Folder layouts can change; the name is portable.",{"type":82,"tag":164,"props":2356,"children":2357},{},[2358,2360,2365,2367,2373,2375,2381],{"type":88,"value":2359},"Clone or refresh ",{"type":82,"tag":112,"props":2361,"children":2363},{"className":2362},[],[2364],{"type":88,"value":48},{"type":88,"value":2366}," under the\nshared upstream root\n(",{"type":82,"tag":112,"props":2368,"children":2370},{"className":2369},[],[2371],{"type":88,"value":2372},"${NUREC_SKILLS_UPSTREAM_ROOT:-${PHYSICAL_AI_SKILL_HUB_UPSTREAM_ROOT:-$HOME\u002F.physical-ai-skill-hub\u002Fupstreams}}\u002Fnurec-skills",{"type":88,"value":2374},").\nDo not scan broad developer workspaces such as ",{"type":82,"tag":112,"props":2376,"children":2378},{"className":2377},[],[2379],{"type":88,"value":2380},"~\u002FCodes",{"type":88,"value":2382}," or reuse\nunrelated old clones.",{"type":82,"tag":164,"props":2384,"children":2385},{},[2386,2391,2393,2399],{"type":82,"tag":112,"props":2387,"children":2389},{"className":2388},[],[2390],{"type":88,"value":56},{"type":88,"value":2392}," covers gated Hugging Face datasets. Do not\nbypass dataset license terms; the user must accept the\n",{"type":82,"tag":112,"props":2394,"children":2396},{"className":2395},[],[2397],{"type":88,"value":2398},"PhysicalAI-*",{"type":88,"value":2400}," gated licenses on Hugging Face and provide a token\nbefore downloading.",{"type":82,"tag":164,"props":2402,"children":2403},{},[2404,2406,2411,2413,2418,2420,2426,2428,2433],{"type":88,"value":2405},"Asset Harvester runs ",{"type":82,"tag":104,"props":2407,"children":2408},{},[2409],{"type":88,"value":2410},"before",{"type":88,"value":2412}," packaging into a USDZ. Do not call\n",{"type":82,"tag":112,"props":2414,"children":2416},{"className":2415},[],[2417],{"type":88,"value":64},{"type":88,"value":2419},"'s ",{"type":82,"tag":112,"props":2421,"children":2423},{"className":2422},[],[2424],{"type":88,"value":2425},"export-external-assets",{"type":88,"value":2427}," on hand-rolled ",{"type":82,"tag":112,"props":2429,"children":2431},{"className":2430},[],[2432],{"type":88,"value":1757},{"type":88,"value":2434}," files unless\nthe user explicitly asks to skip Asset Harvester.",{"type":82,"tag":164,"props":2436,"children":2437},{},[2438,2440,2445,2447,2452,2454,2459],{"type":88,"value":2439},"For artifact cleanup, prefer the built-in ",{"type":82,"tag":112,"props":2441,"children":2443},{"className":2442},[],[2444],{"type":88,"value":1363},{"type":88,"value":2446}," path in\n",{"type":82,"tag":112,"props":2448,"children":2450},{"className":2449},[],[2451],{"type":88,"value":64},{"type":88,"value":2453},". Route to the standalone ",{"type":82,"tag":112,"props":2455,"children":2457},{"className":2456},[],[2458],{"type":88,"value":72},{"type":88,"value":2460}," only when the user\nneeds the public code\u002Fmodel card, paired evaluation, fine-tuning,\nor fixes on previously rendered frames.",{"type":82,"tag":164,"props":2462,"children":2463},{},[2464,2466,2472],{"type":88,"value":2465},"Do not invent NRE \u002F NCore \u002F DiffusionHarmonizer commands from\nmemory. Re-read the upstream sibling skill — versions move fast\n(NRE ",{"type":82,"tag":112,"props":2467,"children":2469},{"className":2468},[],[2470],{"type":88,"value":2471},"release_26.04",{"type":88,"value":2473}," is the current pinned tag).",{"type":82,"tag":164,"props":2475,"children":2476},{},[2477,2479,2484],{"type":88,"value":2478},"This router does not deploy infrastructure. Route AKS \u002F OSMO \u002F\nNIM Operator setup to\n",{"type":82,"tag":112,"props":2480,"children":2482},{"className":2481},[],[2483],{"type":88,"value":243},{"type":88,"value":220},{"type":82,"tag":91,"props":2486,"children":2488},{"id":2487},"limitations",[2489],{"type":88,"value":2490},"Limitations",{"type":82,"tag":160,"props":2492,"children":2493},{},[2494,2504,2529,2557,2577,2587],{"type":82,"tag":164,"props":2495,"children":2496},{},[2497,2502],{"type":82,"tag":104,"props":2498,"children":2499},{},[2500],{"type":88,"value":2501},"Router only.",{"type":88,"value":2503}," This skill never executes mutating NuRec commands.\nAll training, rendering, conversion, and harmonization happens in\nupstream sibling skills.",{"type":82,"tag":164,"props":2505,"children":2506},{},[2507,2512,2514,2519,2521,2527],{"type":82,"tag":104,"props":2508,"children":2509},{},[2510],{"type":88,"value":2511},"Upstream-pinned.",{"type":88,"value":2513}," Recipes live in\n",{"type":82,"tag":112,"props":2515,"children":2517},{"className":2516},[],[2518],{"type":88,"value":48},{"type":88,"value":2520},", which evolves outside\nthis repo. Stale clones can drift; always ",{"type":82,"tag":112,"props":2522,"children":2524},{"className":2523},[],[2525],{"type":88,"value":2526},"git pull",{"type":88,"value":2528}," the upstream\nbefore relying on a sibling skill.",{"type":82,"tag":164,"props":2530,"children":2531},{},[2532,2537,2538,2543,2544,2549,2550,2555],{"type":82,"tag":104,"props":2533,"children":2534},{},[2535],{"type":88,"value":2536},"Gated content.",{"type":88,"value":1357},{"type":82,"tag":112,"props":2539,"children":2541},{"className":2540},[],[2542],{"type":88,"value":583},{"type":88,"value":132},{"type":82,"tag":112,"props":2545,"children":2547},{"className":2546},[],[2548],{"type":88,"value":590},{"type":88,"value":514},{"type":82,"tag":112,"props":2551,"children":2553},{"className":2552},[],[2554],{"type":88,"value":598},{"type":88,"value":2556}," require the user to accept license\nterms on Hugging Face first. The router cannot bypass this.",{"type":82,"tag":164,"props":2558,"children":2559},{},[2560,2565,2567,2576],{"type":82,"tag":104,"props":2561,"children":2562},{},[2563],{"type":88,"value":2564},"Heavy footprint.",{"type":88,"value":2566}," A complete NuRec workflow can leave 150 GB+\non disk. See ",{"type":82,"tag":461,"props":2568,"children":2570},{"href":2569},"references\u002Fteardown.md",[2571],{"type":82,"tag":112,"props":2572,"children":2574},{"className":2573},[],[2575],{"type":88,"value":2569},{"type":88,"value":220},{"type":82,"tag":164,"props":2578,"children":2579},{},[2580,2585],{"type":82,"tag":104,"props":2581,"children":2582},{},[2583],{"type":88,"value":2584},"NVIDIA-only stack.",{"type":88,"value":2586}," Requires an NVIDIA GPU plus the NVIDIA\nContainer Toolkit. AMD \u002F Intel \u002F Apple Silicon are not supported.",{"type":82,"tag":164,"props":2588,"children":2589},{},[2590,2595,2597,2602],{"type":82,"tag":104,"props":2591,"children":2592},{},[2593],{"type":88,"value":2594},"Not a SimReady pipeline.",{"type":88,"value":2596}," NuRec produces a renderable USDZ from\na recording; SimReady packaging of CAD or source meshes is a\ndifferent pipeline (see ",{"type":82,"tag":112,"props":2598,"children":2600},{"className":2599},[],[2601],{"type":88,"value":218},{"type":88,"value":542},{"type":82,"tag":91,"props":2604,"children":2606},{"id":2605},"troubleshooting",[2607],{"type":88,"value":2608},"Troubleshooting",{"type":82,"tag":1022,"props":2610,"children":2611},{},[2612,2633],{"type":82,"tag":1026,"props":2613,"children":2614},{},[2615],{"type":82,"tag":1030,"props":2616,"children":2617},{},[2618,2623,2628],{"type":82,"tag":1034,"props":2619,"children":2620},{},[2621],{"type":88,"value":2622},"Error \u002F symptom",{"type":82,"tag":1034,"props":2624,"children":2625},{},[2626],{"type":88,"value":2627},"Likely cause",{"type":82,"tag":1034,"props":2629,"children":2630},{},[2631],{"type":88,"value":2632},"Solution",{"type":82,"tag":1045,"props":2634,"children":2635},{},[2636,2663,2724,2786,2816,2860,2911,2944,3001],{"type":82,"tag":1030,"props":2637,"children":2638},{},[2639,2649,2654],{"type":82,"tag":1052,"props":2640,"children":2641},{},[2642,2647],{"type":82,"tag":112,"props":2643,"children":2645},{"className":2644},[],[2646],{"type":88,"value":179},{"type":88,"value":2648}," clone missing or empty",{"type":82,"tag":1052,"props":2650,"children":2651},{},[2652],{"type":88,"value":2653},"Upstream not fetched yet",{"type":82,"tag":1052,"props":2655,"children":2656},{},[2657,2659],{"type":88,"value":2658},"Run the clone block in ",{"type":82,"tag":461,"props":2660,"children":2661},{"href":463},[2662],{"type":88,"value":466},{"type":82,"tag":1030,"props":2664,"children":2665},{},[2666,2691,2703],{"type":82,"tag":1052,"props":2667,"children":2668},{},[2669,2675,2676,2682,2684,2689],{"type":82,"tag":112,"props":2670,"children":2672},{"className":2671},[],[2673],{"type":88,"value":2674},"403",{"type":88,"value":1880},{"type":82,"tag":112,"props":2677,"children":2679},{"className":2678},[],[2680],{"type":88,"value":2681},"401",{"type":88,"value":2683}," pulling ",{"type":82,"tag":112,"props":2685,"children":2687},{"className":2686},[],[2688],{"type":88,"value":583},{"type":88,"value":2690}," from HF",{"type":82,"tag":1052,"props":2692,"children":2693},{},[2694,2696,2701],{"type":88,"value":2695},"Gated license not accepted, or ",{"type":82,"tag":112,"props":2697,"children":2699},{"className":2698},[],[2700],{"type":88,"value":575},{"type":88,"value":2702}," unset \u002F wrong scope",{"type":82,"tag":1052,"props":2704,"children":2705},{},[2706,2708,2714,2716,2722],{"type":88,"value":2707},"Accept the gated license on Hugging Face, then ",{"type":82,"tag":112,"props":2709,"children":2711},{"className":2710},[],[2712],{"type":88,"value":2713},"hf auth login",{"type":88,"value":2715}," with a token that has ",{"type":82,"tag":112,"props":2717,"children":2719},{"className":2718},[],[2720],{"type":88,"value":2721},"read",{"type":88,"value":2723}," access",{"type":82,"tag":1030,"props":2725,"children":2726},{},[2727,2744,2754],{"type":82,"tag":1052,"props":2728,"children":2729},{},[2730,2736,2738],{"type":82,"tag":112,"props":2731,"children":2733},{"className":2732},[],[2734],{"type":88,"value":2735},"denied: requested access to the resource is denied",{"type":88,"value":2737}," from ",{"type":82,"tag":112,"props":2739,"children":2741},{"className":2740},[],[2742],{"type":88,"value":2743},"nvcr.io\u002Fnvidia\u002Fnre\u002F*",{"type":82,"tag":1052,"props":2745,"children":2746},{},[2747,2749],{"type":88,"value":2748},"Missing or expired ",{"type":82,"tag":112,"props":2750,"children":2752},{"className":2751},[],[2753],{"type":88,"value":558},{"type":82,"tag":1052,"props":2755,"children":2756},{},[2757,2763,2764,2770,2771,2776,2778,2784],{"type":82,"tag":112,"props":2758,"children":2760},{"className":2759},[],[2761],{"type":88,"value":2762},"docker login nvcr.io",{"type":88,"value":617},{"type":82,"tag":112,"props":2765,"children":2767},{"className":2766},[],[2768],{"type":88,"value":2769},"$oauthtoken",{"type":88,"value":1437},{"type":82,"tag":112,"props":2772,"children":2774},{"className":2773},[],[2775],{"type":88,"value":558},{"type":88,"value":2777},"; rotate the key at ",{"type":82,"tag":112,"props":2779,"children":2781},{"className":2780},[],[2782],{"type":88,"value":2783},"org.ngc.nvidia.com\u002Fsetup\u002Fapi-key",{"type":88,"value":2785}," if needed",{"type":82,"tag":1030,"props":2787,"children":2788},{},[2789,2794,2799],{"type":82,"tag":1052,"props":2790,"children":2791},{},[2792],{"type":88,"value":2793},"NRE refuses to load a clip (\"not valid NCore V4\")",{"type":82,"tag":1052,"props":2795,"children":2796},{},[2797],{"type":88,"value":2798},"Recording was not converted",{"type":82,"tag":1052,"props":2800,"children":2801},{},[2802,2804,2809,2811],{"type":88,"value":2803},"Run the ",{"type":82,"tag":112,"props":2805,"children":2807},{"className":2806},[],[2808],{"type":88,"value":60},{"type":88,"value":2810}," skill before invoking ",{"type":82,"tag":112,"props":2812,"children":2814},{"className":2813},[],[2815],{"type":88,"value":64},{"type":82,"tag":1030,"props":2817,"children":2818},{},[2819,2829,2834],{"type":82,"tag":1052,"props":2820,"children":2821},{},[2822,2827],{"type":82,"tag":112,"props":2823,"children":2825},{"className":2824},[],[2826],{"type":88,"value":417},{"type":88,"value":2828}," cold-start latency dominates a Python loop",{"type":82,"tag":1052,"props":2830,"children":2831},{},[2832],{"type":88,"value":2833},"One-shot Docker invocation per render",{"type":82,"tag":1052,"props":2835,"children":2836},{},[2837,2839,2844,2846,2851,2853,2858],{"type":88,"value":2838},"Use the ",{"type":82,"tag":112,"props":2840,"children":2842},{"className":2841},[],[2843],{"type":88,"value":64},{"type":88,"value":2845}," warm ",{"type":82,"tag":112,"props":2847,"children":2849},{"className":2848},[],[2850],{"type":88,"value":417},{"type":88,"value":2852}," + thin Python client (",{"type":82,"tag":112,"props":2854,"children":2856},{"className":2855},[],[2857],{"type":88,"value":445},{"type":88,"value":2859},") recipe",{"type":82,"tag":1030,"props":2861,"children":2862},{},[2863,2881,2892],{"type":82,"tag":1052,"props":2864,"children":2865},{},[2866,2868,2873,2875],{"type":88,"value":2867},"Output files are owned by ",{"type":82,"tag":112,"props":2869,"children":2871},{"className":2870},[],[2872],{"type":88,"value":79},{"type":88,"value":2874}," after a ",{"type":82,"tag":112,"props":2876,"children":2878},{"className":2877},[],[2879],{"type":88,"value":2880},"docker run",{"type":82,"tag":1052,"props":2882,"children":2883},{},[2884,2890],{"type":82,"tag":112,"props":2885,"children":2887},{"className":2886},[],[2888],{"type":88,"value":2889},"-u $(id -u):$(id -g)",{"type":88,"value":2891}," was missing",{"type":82,"tag":1052,"props":2893,"children":2894},{},[2895,2901,2903,2909],{"type":82,"tag":112,"props":2896,"children":2898},{"className":2897},[],[2899],{"type":88,"value":2900},"sudo chown -R \"$(id -u):$(id -g)\" \u003Coutput_dir>",{"type":88,"value":2902},"; add the ",{"type":82,"tag":112,"props":2904,"children":2906},{"className":2905},[],[2907],{"type":88,"value":2908},"-u",{"type":88,"value":2910}," flag next time",{"type":82,"tag":1030,"props":2912,"children":2913},{},[2914,2919,2924],{"type":82,"tag":1052,"props":2915,"children":2916},{},[2917],{"type":88,"value":2918},"Frames have ghosting \u002F floaters \u002F flicker after rendering",{"type":82,"tag":1052,"props":2920,"children":2921},{},[2922],{"type":88,"value":2923},"Inline cleanup not enabled",{"type":82,"tag":1052,"props":2925,"children":2926},{},[2927,2929,2935,2937,2942],{"type":88,"value":2928},"Re-render with ",{"type":82,"tag":112,"props":2930,"children":2932},{"className":2931},[],[2933],{"type":88,"value":2934},"nre --enable-difix",{"type":88,"value":2936},", or post-process with ",{"type":82,"tag":112,"props":2938,"children":2940},{"className":2939},[],[2941],{"type":88,"value":72},{"type":88,"value":2943}," (DiffusionHarmonizer)",{"type":82,"tag":1030,"props":2945,"children":2946},{},[2947,2968,2973],{"type":82,"tag":1052,"props":2948,"children":2949},{},[2950,2952,2958,2960,2966],{"type":88,"value":2951},"Stale skill names (",{"type":82,"tag":112,"props":2953,"children":2955},{"className":2954},[],[2956],{"type":88,"value":2957},"ncore-data-conversion",{"type":88,"value":2959},", old ",{"type":82,"tag":112,"props":2961,"children":2963},{"className":2962},[],[2964],{"type":88,"value":2965},"nvidia\u002FFixer",{"type":88,"value":2967},") in agent output",{"type":82,"tag":1052,"props":2969,"children":2970},{},[2971],{"type":88,"value":2972},"Out-of-date cached skill",{"type":82,"tag":1052,"props":2974,"children":2975},{},[2976,2978,2983,2985,2990,2992],{"type":88,"value":2977},"Update references to ",{"type":82,"tag":112,"props":2979,"children":2981},{"className":2980},[],[2982],{"type":88,"value":60},{"type":88,"value":2984}," and ",{"type":82,"tag":112,"props":2986,"children":2988},{"className":2987},[],[2989],{"type":88,"value":72},{"type":88,"value":2991}," (DiffusionHarmonizer); see ",{"type":82,"tag":461,"props":2993,"children":2995},{"href":2994},"references\u002Fmaintenance.md",[2996],{"type":82,"tag":112,"props":2997,"children":2999},{"className":2998},[],[3000],{"type":88,"value":2994},{"type":82,"tag":1030,"props":3002,"children":3003},{},[3004,3017,3022],{"type":82,"tag":1052,"props":3005,"children":3006},{},[3007,3009,3015],{"type":88,"value":3008},"Bash anti-pattern ",{"type":82,"tag":112,"props":3010,"children":3012},{"className":3011},[],[3013],{"type":88,"value":3014},"${HF_TOKEN:+yes}${HF_TOKEN:-no}",{"type":88,"value":3016}," echoed token value",{"type":82,"tag":1052,"props":3018,"children":3019},{},[3020],{"type":88,"value":3021},"Misuse of bash parameter expansion",{"type":82,"tag":1052,"props":3023,"children":3024},{},[3025,3027,3033,3035,3043],{"type":88,"value":3026},"Rotate the token; use ",{"type":82,"tag":112,"props":3028,"children":3030},{"className":3029},[],[3031],{"type":88,"value":3032},"hf auth whoami",{"type":88,"value":3034}," or length-only checks (see ",{"type":82,"tag":461,"props":3036,"children":3037},{"href":861},[3038],{"type":82,"tag":112,"props":3039,"children":3041},{"className":3040},[],[3042],{"type":88,"value":861},{"type":88,"value":1218},{"type":82,"tag":91,"props":3045,"children":3047},{"id":3046},"cross-skill-teardown",[3048],{"type":88,"value":3049},"Cross-skill teardown",{"type":82,"tag":98,"props":3051,"children":3052},{},[3053,3055,3060,3062,3068,3070,3078],{"type":88,"value":3054},"A complete NuRec workflow can leave ",{"type":82,"tag":104,"props":3056,"children":3057},{},[3058],{"type":88,"value":3059},"150 GB+",{"type":88,"value":3061}," on disk between\ncontainer images, model weights, code clones, conda envs, and output\ndirectories. Each sibling skill has its own dedicated ",{"type":82,"tag":112,"props":3063,"children":3065},{"className":3064},[],[3066],{"type":88,"value":3067},"Teardown",{"type":88,"value":3069},"\nsection — read them in the order documented in\n",{"type":82,"tag":461,"props":3071,"children":3072},{"href":2569},[3073],{"type":82,"tag":112,"props":3074,"children":3076},{"className":3075},[],[3077],{"type":88,"value":2569},{"type":88,"value":3079}," when the user no\nlonger needs the workflow.",{"type":82,"tag":91,"props":3081,"children":3083},{"id":3082},"keeping-this-router-up-to-date",[3084],{"type":88,"value":3085},"Keeping this router up to date",{"type":82,"tag":98,"props":3087,"children":3088},{},[3089,3091,3099,3101,3106,3108,3115],{"type":88,"value":3090},"Procedure for adding new sibling skills, renames, or upstream URL\nchanges lives in ",{"type":82,"tag":461,"props":3092,"children":3093},{"href":2994},[3094],{"type":82,"tag":112,"props":3095,"children":3097},{"className":3096},[],[3098],{"type":88,"value":2994},{"type":88,"value":3100},".\nTreat the upstream ",{"type":82,"tag":112,"props":3102,"children":3104},{"className":3103},[],[3105],{"type":88,"value":53},{"type":88,"value":3107}," at\n",{"type":82,"tag":461,"props":3109,"children":3113},{"href":3110,"rel":3111},"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fnurec-skills\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002FSKILL.md",[3112],"nofollow",[3114],{"type":88,"value":3110},{"type":88,"value":3116},"\nas authoritative; this skill mirrors only the picker tables, the\nworkflow ordering, and the upstream fetch recipe.",{"type":82,"tag":3118,"props":3119,"children":3120},"style",{},[3121],{"type":88,"value":3122},"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":3124,"total":3283},[3125,3143,3161,3172,3184,3198,3211,3225,3238,3249,3263,3272],{"slug":3126,"name":3126,"fn":3127,"description":3128,"org":3129,"tags":3130,"stars":3140,"repoUrl":3141,"updatedAt":3142},"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},[3131,3134,3137],{"name":3132,"slug":3133,"type":15},"Documentation","documentation",{"name":3135,"slug":3136,"type":15},"MCP","mcp",{"name":3138,"slug":3139,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":3144,"name":3144,"fn":3145,"description":3146,"org":3147,"tags":3148,"stars":3158,"repoUrl":3159,"updatedAt":3160},"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},[3149,3152,3155],{"name":3150,"slug":3151,"type":15},"Containers","containers",{"name":3153,"slug":3154,"type":15},"Deployment","deployment",{"name":3156,"slug":3157,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":3162,"name":3162,"fn":3163,"description":3164,"org":3165,"tags":3166,"stars":3158,"repoUrl":3159,"updatedAt":3171},"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},[3167,3170],{"name":3168,"slug":3169,"type":15},"CI\u002FCD","ci-cd",{"name":3153,"slug":3154,"type":15},"2026-07-14T05:25:59.97109",{"slug":3173,"name":3173,"fn":3174,"description":3175,"org":3176,"tags":3177,"stars":3158,"repoUrl":3159,"updatedAt":3183},"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},[3178,3179,3180],{"name":3168,"slug":3169,"type":15},{"name":3153,"slug":3154,"type":15},{"name":3181,"slug":3182,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":3185,"name":3185,"fn":3186,"description":3187,"org":3188,"tags":3189,"stars":3158,"repoUrl":3159,"updatedAt":3197},"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},[3190,3193,3194],{"name":3191,"slug":3192,"type":15},"Debugging","debugging",{"name":3181,"slug":3182,"type":15},{"name":3195,"slug":3196,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":3199,"name":3199,"fn":3200,"description":3201,"org":3202,"tags":3203,"stars":3158,"repoUrl":3159,"updatedAt":3210},"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},[3204,3207],{"name":3205,"slug":3206,"type":15},"Best Practices","best-practices",{"name":3208,"slug":3209,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":3212,"name":3212,"fn":3213,"description":3214,"org":3215,"tags":3216,"stars":3158,"repoUrl":3159,"updatedAt":3224},"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},[3217,3220,3223],{"name":3218,"slug":3219,"type":15},"Machine Learning","machine-learning",{"name":3221,"slug":3222,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":3226,"name":3226,"fn":3227,"description":3228,"org":3229,"tags":3230,"stars":3158,"repoUrl":3159,"updatedAt":3237},"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},[3231,3234],{"name":3232,"slug":3233,"type":15},"QA","qa",{"name":3235,"slug":3236,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":3239,"name":3239,"fn":3240,"description":3241,"org":3242,"tags":3243,"stars":3158,"repoUrl":3159,"updatedAt":3248},"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},[3244,3245],{"name":3153,"slug":3154,"type":15},{"name":3246,"slug":3247,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":3250,"name":3250,"fn":3251,"description":3252,"org":3253,"tags":3254,"stars":3158,"repoUrl":3159,"updatedAt":3262},"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},[3255,3258,3259],{"name":3256,"slug":3257,"type":15},"Code Review","code-review",{"name":3181,"slug":3182,"type":15},{"name":3260,"slug":3261,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":3264,"name":3264,"fn":3265,"description":3266,"org":3267,"tags":3268,"stars":3158,"repoUrl":3159,"updatedAt":3271},"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},[3269,3270],{"name":3232,"slug":3233,"type":15},{"name":3235,"slug":3236,"type":15},"2026-07-14T05:25:54.928983",{"slug":3273,"name":3273,"fn":3274,"description":3275,"org":3276,"tags":3277,"stars":3158,"repoUrl":3159,"updatedAt":3282},"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},[3278,3281],{"name":3279,"slug":3280,"type":15},"Automation","automation",{"name":3168,"slug":3169,"type":15},"2026-07-30T05:29:03.275638",496,{"items":3285,"total":3379},[3286,3303,3313,3327,3337,3352,3367],{"slug":3287,"name":3287,"fn":3288,"description":3289,"org":3290,"tags":3291,"stars":23,"repoUrl":24,"updatedAt":3302},"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},[3292,3295,3298,3299],{"name":3293,"slug":3294,"type":15},"Data Analysis","data-analysis",{"name":3296,"slug":3297,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":3300,"slug":3301,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":3304,"name":3304,"fn":3305,"description":3306,"org":3307,"tags":3308,"stars":23,"repoUrl":24,"updatedAt":3312},"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},[3309,3310,3311],{"name":3153,"slug":3154,"type":15},{"name":3246,"slug":3247,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":3314,"name":3314,"fn":3315,"description":3316,"org":3317,"tags":3318,"stars":23,"repoUrl":24,"updatedAt":3326},"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},[3319,3322,3323],{"name":3320,"slug":3321,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":3324,"slug":3325,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":3328,"name":3328,"fn":3329,"description":3330,"org":3331,"tags":3332,"stars":23,"repoUrl":24,"updatedAt":3336},"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},[3333,3334,3335],{"name":3293,"slug":3294,"type":15},{"name":9,"slug":8,"type":15},{"name":3235,"slug":3236,"type":15},"2026-07-17T05:29:03.913266",{"slug":3338,"name":3338,"fn":3339,"description":3340,"org":3341,"tags":3342,"stars":23,"repoUrl":24,"updatedAt":3351},"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},[3343,3344,3347,3348],{"name":3279,"slug":3280,"type":15},{"name":3345,"slug":3346,"type":15},"Imaging","imaging",{"name":9,"slug":8,"type":15},{"name":3349,"slug":3350,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":3353,"name":3353,"fn":3354,"description":3355,"org":3356,"tags":3357,"stars":23,"repoUrl":24,"updatedAt":3366},"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},[3358,3359,3362,3363],{"name":3153,"slug":3154,"type":15},{"name":3360,"slug":3361,"type":15},"Docker","docker",{"name":9,"slug":8,"type":15},{"name":3364,"slug":3365,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":3368,"name":3368,"fn":3369,"description":3370,"org":3371,"tags":3372,"stars":23,"repoUrl":24,"updatedAt":3378},"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},[3373,3374,3377],{"name":9,"slug":8,"type":15},{"name":3375,"slug":3376,"type":15},"Quantum Computing","quantum-computing",{"name":18,"slug":19,"type":15},"2026-07-14T05:26:58.898253",305]