[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78Z9X4KAMGF3ZKT8JFN74Q9":3,"skill-same:01M78Z9X4KAMGF3ZKT8JFN74Q9":67},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M78Z9X4KAMGF3ZKT8JFN74Q9","distributed-accelerators",null,"Choose, configure, validate, and troubleshoot Lightning Trainer and Fabric accelerators, devices, precision, DDP, FSDP, DeepSpeed, model parallel, cluster, and hardware backends.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: distributed-accelerators\ndescription: \"Choose, configure, validate, and troubleshoot Lightning Trainer\n  and Fabric accelerators, devices, precision, DDP, FSDP, DeepSpeed, model\n  parallel, cluster, and hardware backends.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# Lightning Distributed Accelerators\n\nUse this sub-skill when the task involves `Trainer` or `Fabric` accelerator\u002Fdevice selection, precision, distributed strategies, FSDP, DeepSpeed, DDP, tensor\u002Fmodel parallel, CUDA\u002FGPU, MPS, TPU\u002FXLA, multi-node clusters, or optional dependency and hardware troubleshooting.\n\n## Route Here\n\n- Choose `accelerator`, `devices`, `num_nodes`, `strategy`, `precision`, plugins, and cluster environment settings for `lightning.pytorch.Trainer` or `lightning.Fabric`.\n- Diagnose GPU visibility, device-count mismatches, CPU fallback, MPS\u002FTPU\u002FXLA limitations, NCCL\u002FGloo launch problems, and optional dependency import failures.\n- Configure `DDPStrategy`, `FSDPStrategy`, `DeepSpeedStrategy`, `ModelParallelStrategy`, precision plugins, or strategy registry entries.\n- Plan large-model training with DDP, FSDP, DeepSpeed ZeRO, activation checkpointing, CPU offload, tensor parallelism, or 2D parallelism.\n\n## Route Elsewhere\n\n- Core `LightningModule`, `LightningDataModule`, callbacks, loggers, checkpoint callbacks, and standard `Trainer.fit` semantics: `..\u002Ftraining-core\u002FSKILL.md`.\n- Manual Fabric loop structure, `fabric.setup`, `fabric.backward`, checkpoint IO, and custom loop ownership: `..\u002Ffabric-expert-loops\u002FSKILL.md`.\n- `LightningCLI`, YAML, environment-variable, and parser syntax for these settings: `..\u002Fcli-configuration\u002FSKILL.md`.\n\n## Start Here\n\n1. Identify whether the user is using `Trainer`, `Fabric`, or CLI config; route to sibling sub-skills only for the non-distributed wiring.\n2. Pick a baseline from `references\u002Fstrategy-guide.md`: start with `strategy=\"auto\"` for one process, `ddp` for normal multi-GPU, `fsdp` for DDP out-of-memory, and `deepspeed` for DeepSpeed-specific ZeRO\u002Foffload workflows.\n3. Cross-check accelerator\u002Fdevice\u002Fprecision support in `references\u002Fprecision-and-devices.md` before recommending GPU, MPS, TPU, FP16, BF16, FP8, or quantization settings.\n4. Validate syntax without launching workers:\n\n```bash\npython scripts\u002Fstrategy_config_check.py --mode trainer --accelerator cpu --devices 1 --strategy auto --precision 32-true\npython scripts\u002Fstrategy_config_check.py --mode fabric --accelerator auto --devices 1 --strategy auto --precision bf16-mixed\n```\n\n5. For failures, use `references\u002Ftroubleshooting.md` and report whether the evidence is import-only, syntax-only, or requires real distributed\u002FGPU validation. This skill was drafted from CPU inspection evidence and does not claim GPU or distributed runtime was validated.\n\n## Reference Map\n\n- `references\u002Fstrategy-guide.md`: decision table, Trainer\u002FFabric configuration patterns, DDP\u002FFSDP\u002FDeepSpeed\u002Fmodel-parallel examples, strategy constructor knobs, and cluster launch notes.\n- `references\u002Fprecision-and-devices.md`: accelerator names, `devices` formats, precision modes, optional precision plugins, and hardware support boundaries.\n- `references\u002Ftroubleshooting.md`: install\u002Fimport, optional dependencies, invalid API usage, backend limitations, launch errors, device mismatch, and workflow-specific misconfigurations.\n- `scripts\u002Fstrategy_config_check.py`: safe public-package validator for import availability and syntactic configuration checks; it does not spawn subprocesses or start distributed workers.\n","# Lightning Distributed Accelerators\n\nUse this sub-skill when the task involves `Trainer` or `Fabric` accelerator\u002Fdevice selection, precision, distributed strategies, FSDP, DeepSpeed, DDP, tensor\u002Fmodel parallel, CUDA\u002FGPU, MPS, TPU\u002FXLA, multi-node clusters, or optional dependency and hardware troubleshooting.\n\n## Route Here\n\n- Choose `accelerator`, `devices`, `num_nodes`, `strategy`, `precision`, plugins, and cluster environment settings for `lightning.pytorch.Trainer` or `lightning.Fabric`.\n- Diagnose GPU visibility, device-count mismatches, CPU fallback, MPS\u002FTPU\u002FXLA limitations, NCCL\u002FGloo launch problems, and optional dependency import failures.\n- Configure `DDPStrategy`, `FSDPStrategy`, `DeepSpeedStrategy`, `ModelParallelStrategy`, precision plugins, or strategy registry entries.\n- Plan large-model training with DDP, FSDP, DeepSpeed ZeRO, activation checkpointing, CPU offload, tensor parallelism, or 2D parallelism.\n\n## Route Elsewhere\n\n- Core `LightningModule`, `LightningDataModule`, callbacks, loggers, checkpoint callbacks, and standard `Trainer.fit` semantics: `..\u002Ftraining-core\u002FSKILL.md`.\n- Manual Fabric loop structure, `fabric.setup`, `fabric.backward`, checkpoint IO, and custom loop ownership: `..\u002Ffabric-expert-loops\u002FSKILL.md`.\n- `LightningCLI`, YAML, environment-variable, and parser syntax for these settings: `..\u002Fcli-configuration\u002FSKILL.md`.\n\n## Start Here\n\n1. Identify whether the user is using `Trainer`, `Fabric`, or CLI config; route to sibling sub-skills only for the non-distributed wiring.\n2. Pick a baseline from `references\u002Fstrategy-guide.md`: start with `strategy=\"auto\"` for one process, `ddp` for normal multi-GPU, `fsdp` for DDP out-of-memory, and `deepspeed` for DeepSpeed-specific ZeRO\u002Foffload workflows.\n3. Cross-check accelerator\u002Fdevice\u002Fprecision support in `references\u002Fprecision-and-devices.md` before recommending GPU, MPS, TPU, FP16, BF16, FP8, or quantization settings.\n4. Validate syntax without launching workers:\n\n```bash\npython scripts\u002Fstrategy_config_check.py --mode trainer --accelerator cpu --devices 1 --strategy auto --precision 32-true\npython scripts\u002Fstrategy_config_check.py --mode fabric --accelerator auto --devices 1 --strategy auto --precision bf16-mixed\n```\n\n5. For failures, use `references\u002Ftroubleshooting.md` and report whether the evidence is import-only, syntax-only, or requires real distributed\u002FGPU validation. This skill was drafted from CPU inspection evidence and does not claim GPU or distributed runtime was validated.\n\n## Reference Map\n\n- `references\u002Fstrategy-guide.md`: decision table, Trainer\u002FFabric configuration patterns, DDP\u002FFSDP\u002FDeepSpeed\u002Fmodel-parallel examples, strategy constructor knobs, and cluster launch notes.\n- `references\u002Fprecision-and-devices.md`: accelerator names, `devices` formats, precision modes, optional precision plugins, and hardware support boundaries.\n- `references\u002Ftroubleshooting.md`: install\u002Fimport, optional dependencies, invalid API usage, backend limitations, launch errors, device mismatch, and workflow-specific misconfigurations.\n- `scripts\u002Fstrategy_config_check.py`: safe public-package validator for import availability and syntactic configuration checks; it does not spawn subprocesses or start distributed workers.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill distributed-accelerators",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill distributed-accelerators",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill distributed-accelerators","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators\u002FSKILL.md",[47,53,57,61],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fprecision-and-devices.md","REFERENCE","参考文档",6165,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators\u002Freferences\u002Fprecision-and-devices.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fstrategy-guide.md",10552,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators\u002Freferences\u002Fstrategy-guide.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Ftroubleshooting.md",8418,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators\u002Freferences\u002Ftroubleshooting.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Fstrategy_config_check.py","SCRIPT","脚本",9047,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flightning\u002Fsub-skills\u002Fdistributed-accelerators\u002Fscripts\u002Fstrategy_config_check.py",[68,75,82,89,96,103],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"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.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"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":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"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":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"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":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"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",[],[]]