[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78VNNG2175KAYES5PY8V42F":3,"skill-same:01M78VNNG2175KAYES5PY8V42F":74},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":32,"install":33,"resources":45},"01M78VNNG2175KAYES5PY8V42F","deep-gcns-torch",null,"Use the DeepGCNs PyTorch operating guide for graph convolution layers, dynamic and dilated KNN blocks, point-cloud classification or segmentation, PPI, OGB\u002FDeeperGCN, and reversible memory-efficient GNN workflows.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"MIT","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: deep-gcns-torch\ndescription: \"Use the DeepGCNs PyTorch operating guide for graph convolution\n  layers, dynamic and dilated KNN blocks, point-cloud classification or\n  segmentation, PPI, OGB\u002FDeeperGCN, and reversible memory-efficient GNN\n  workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# DeepGCNs Torch\n\nUse this router when a task names DeepGCNs, DeeperGCN, `deep_gcns_torch`,\n`gcn_lib`, GENConv, dynamic KNN graph blocks, ModelNet40\u002FS3DIS\u002FPartNet point\nclouds, PPI, OGB graph benchmarks, RevGNN, or RevGAT.\n\n## Operating boundary\n\nThis is a self-contained guide distilled from a historical PyTorch repository.\nIt does not download datasets, fetch external checkpoints, launch long\ntraining, start distributed jobs, or open visualization windows. First stage\ncaller-owned data and checkpoints, then follow the relevant workflow. Read\n[repository provenance](references\u002Frepo-provenance.md) when deciding whether a\nsource checkout has drifted, and read [troubleshooting](references\u002Ftroubleshooting.md)\nwhen installation or version behavior is unclear.\n\n## Minimal environment check\n\nThe source repository has no `pyproject.toml`, `setup.py`, or console entry\npoint. Use a coherent PyTorch\u002FPyG stack rather than the obsolete repository\ninstaller. At minimum, install PyTorch, PyTorch Geometric, matching\n`torch-scatter` and `torch-cluster` wheels, and add `ogb`, h5py, scikit-learn,\nor tensorboard only for the selected workflows. Run the bundled checks before\nusing a real model:\n\n```bash\npython scripts\u002Fcheck_env.py\npython scripts\u002Fcheckpoint_roundtrip.py\n```\n\nFor a CUDA task, use `python scripts\u002Fcheck_env.py --cuda`; CUDA availability is\nnot proof of dataset-scale memory or benchmark reproduction. Match every\ncompiled PyG extension to the installed PyTorch and CPU\u002FCUDA build. Do not run\nthe old one-shot installer: it mutates environments and pins obsolete CUDA\n10.2-era packages.\n\n## Route by the user's task\n\n- **Layer or architecture API** — sparse\u002Fdense `GraphConv`, `DynConv`,\n  `GENConv`, aggregators, KNN\u002Fdilation, blocks, or reversible primitives:\n  read [graph-layers](sub-skills\u002Fgraph-layers\u002FSKILL.md).\n- **Point clouds** — ModelNet40, S3DIS dense\u002Fsparse, PartNet, segmentation\n  layouts, point-cloud checkpoints, or optional OBJ visualization: read\n  [point-cloud-workflows](sub-skills\u002Fpoint-cloud-workflows\u002FSKILL.md).\n- **OGB\u002FDeeperGCN** — `ogbn-*`, `ogbg-*`, `ogbl-collab`, graph pooling,\n  partitions, atom\u002Fbond features, RevGNN, or DGL RevGAT: read\n  [ogb-workflows](sub-skills\u002Fogb-workflows\u002FSKILL.md).\n- **PPI** — PPI multilabel node classification, F1 metrics, or PPI checkpoint\n  alignment: read [ppi-workflows](sub-skills\u002Fppi-workflows\u002FSKILL.md).\n\nIf a request crosses routes, start here, state the data layout and backend,\nthen read each owning sub-skill. Generic layer semantics belong to\n`graph-layers`; task-specific data and metrics remain with their workflow.\n\n## Fast safety gates\n\n1. Identify CPU versus CUDA, PyG extension compatibility, data root, checkpoint\n   availability, and expected metric before running anything.\n2. Use the owning sub-skill's bundled synthetic helper first. These helpers are\n   offline, deterministic, and do not import the original checkout.\n3. Never treat a CPU shape smoke as proof of GPU memory, a historical benchmark\n   number, or a dataset split result.\n4. Keep the source task's working-directory import quirks and external download\n   requirements explicit; do not turn them into hidden runtime dependencies.\n5. For a changed source revision or package API, compare provenance and use a\n   refresh workflow rather than silently applying stale claims.\n\n## Shared references and checks\n\n- [API overview](references\u002Fapi-overview.md) summarizes shared tensor, module,\n  and utility contracts before choosing a focused route.\n- [Troubleshooting](references\u002Ftroubleshooting.md) covers cross-cutting install,\n  import, data, checkpoint, backend, and version failures.\n- [Environment check](scripts\u002Fcheck_env.py) is a safe dependency\u002Fbackend probe.\n- [Checkpoint roundtrip](scripts\u002Fcheckpoint_roundtrip.py) verifies basic PyTorch\n  checkpoint serialization without touching caller data.\n","# DeepGCNs Torch\n\nUse this router when a task names DeepGCNs, DeeperGCN, `deep_gcns_torch`,\n`gcn_lib`, GENConv, dynamic KNN graph blocks, ModelNet40\u002FS3DIS\u002FPartNet point\nclouds, PPI, OGB graph benchmarks, RevGNN, or RevGAT.\n\n## Operating boundary\n\nThis is a self-contained guide distilled from a historical PyTorch repository.\nIt does not download datasets, fetch external checkpoints, launch long\ntraining, start distributed jobs, or open visualization windows. First stage\ncaller-owned data and checkpoints, then follow the relevant workflow. Read\n[repository provenance](references\u002Frepo-provenance.md) when deciding whether a\nsource checkout has drifted, and read [troubleshooting](references\u002Ftroubleshooting.md)\nwhen installation or version behavior is unclear.\n\n## Minimal environment check\n\nThe source repository has no `pyproject.toml`, `setup.py`, or console entry\npoint. Use a coherent PyTorch\u002FPyG stack rather than the obsolete repository\ninstaller. At minimum, install PyTorch, PyTorch Geometric, matching\n`torch-scatter` and `torch-cluster` wheels, and add `ogb`, h5py, scikit-learn,\nor tensorboard only for the selected workflows. Run the bundled checks before\nusing a real model:\n\n```bash\npython scripts\u002Fcheck_env.py\npython scripts\u002Fcheckpoint_roundtrip.py\n```\n\nFor a CUDA task, use `python scripts\u002Fcheck_env.py --cuda`; CUDA availability is\nnot proof of dataset-scale memory or benchmark reproduction. Match every\ncompiled PyG extension to the installed PyTorch and CPU\u002FCUDA build. Do not run\nthe old one-shot installer: it mutates environments and pins obsolete CUDA\n10.2-era packages.\n\n## Route by the user's task\n\n- **Layer or architecture API** — sparse\u002Fdense `GraphConv`, `DynConv`,\n  `GENConv`, aggregators, KNN\u002Fdilation, blocks, or reversible primitives:\n  read [graph-layers](sub-skills\u002Fgraph-layers\u002FSKILL.md).\n- **Point clouds** — ModelNet40, S3DIS dense\u002Fsparse, PartNet, segmentation\n  layouts, point-cloud checkpoints, or optional OBJ visualization: read\n  [point-cloud-workflows](sub-skills\u002Fpoint-cloud-workflows\u002FSKILL.md).\n- **OGB\u002FDeeperGCN** — `ogbn-*`, `ogbg-*`, `ogbl-collab`, graph pooling,\n  partitions, atom\u002Fbond features, RevGNN, or DGL RevGAT: read\n  [ogb-workflows](sub-skills\u002Fogb-workflows\u002FSKILL.md).\n- **PPI** — PPI multilabel node classification, F1 metrics, or PPI checkpoint\n  alignment: read [ppi-workflows](sub-skills\u002Fppi-workflows\u002FSKILL.md).\n\nIf a request crosses routes, start here, state the data layout and backend,\nthen read each owning sub-skill. Generic layer semantics belong to\n`graph-layers`; task-specific data and metrics remain with their workflow.\n\n## Fast safety gates\n\n1. Identify CPU versus CUDA, PyG extension compatibility, data root, checkpoint\n   availability, and expected metric before running anything.\n2. Use the owning sub-skill's bundled synthetic helper first. These helpers are\n   offline, deterministic, and do not import the original checkout.\n3. Never treat a CPU shape smoke as proof of GPU memory, a historical benchmark\n   number, or a dataset split result.\n4. Keep the source task's working-directory import quirks and external download\n   requirements explicit; do not turn them into hidden runtime dependencies.\n5. For a changed source revision or package API, compare provenance and use a\n   refresh workflow rather than silently applying stale claims.\n\n## Shared references and checks\n\n- [API overview](references\u002Fapi-overview.md) summarizes shared tensor, module,\n  and utility contracts before choosing a focused route.\n- [Troubleshooting](references\u002Ftroubleshooting.md) covers cross-cutting install,\n  import, data, checkpoint, backend, and version failures.\n- [Environment check](scripts\u002Fcheck_env.py) is a safe dependency\u002Fbackend probe.\n- [Checkpoint roundtrip](scripts\u002Fcheckpoint_roundtrip.py) verifies basic PyTorch\n  checkpoint serialization without touching caller data.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":25},[35,38,41],{"tool":36,"command":37},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill deep-gcns-torch",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill deep-gcns-torch",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill deep-gcns-torch","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002FSKILL.md",[46,52,56,60,64,70],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-overview.md","REFERENCE","参考文档",2551,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002Freferences\u002Fapi-overview.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Frepo-provenance.md",1827,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002Freferences\u002Frepo-provenance.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Frepo-routing-metadata.json",509,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002Freferences\u002Frepo-routing-metadata.json",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Ftroubleshooting.md",3590,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002Freferences\u002Ftroubleshooting.md",{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fcheck_env.py","SCRIPT","脚本",2561,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002Fscripts\u002Fcheck_env.py",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Fcheckpoint_roundtrip.py",1712,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-gcns-torch\u002Fscripts\u002Fcheckpoint_roundtrip.py",[75,83,90,97,104,111],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":79,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]