[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M790CVYFNM5CWG0SB4AH85AS":3,"skill-same:01M790CVYFNM5CWG0SB4AH85AS":70},{"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},"01M790CVYFNM5CWG0SB4AH85AS","inference-workflow",null,"Adapt mixtral-offloading's notebook evidence into safe local offloaded Mixtral inference 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\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: inference-workflow\ndescription: \"Adapt mixtral-offloading's notebook evidence into safe local\n  offloaded Mixtral inference workflows.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: MIT\n---\n\n# Inference Workflow\n\nUse this sub-skill when the task is to run, script, adapt, or troubleshoot the\nMixtral offloading demo workflow: building an offloaded `MixtralForCausalLM`\nfrom quantized safetensors, choosing offload sizes, and generating chat text.\n\n## Read this when\n\n- The user asks to turn the demo notebook into a Python script.\n- The task mentions `OffloadConfig`, `QuantConfig`, `build_model`,\n  `offload_per_layer`, or a quantized Mixtral state directory.\n- The user needs to balance GPU VRAM, system RAM, and generation speed.\n- A local run fails before or during model construction or generation.\n\n## Route map\n\n1. Read [references\u002Fworkflow.md](references\u002Fworkflow.md) for the full\n   notebook-to-script recipe, validation gates, and generation-loop outline.\n2. Read [references\u002Fconfiguration.md](references\u002Fconfiguration.md) when the task\n   is about offload sizing, HQQ quantization settings, or memory trade-offs.\n3. Use [scripts\u002Fcreate_offload_config.py](scripts\u002Fcreate_offload_config.py) to\n   compute `main_size`, `offload_size`, and `buffer_size` from layer\u002Fexpert\n   counts without importing the repository.\n4. Use [scripts\u002Frender_generation_skeleton.py](scripts\u002Frender_generation_skeleton.py)\n   to render a safe starter script that shows where to call `build_model`\n   without downloading weights or executing the model.\n5. Read [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for\n   state-path, CUDA, memory, HQQ config, tokenizer, and generation-cache errors.\n\n## What this sub-skill owns\n\n- Runtime prerequisites specific to offloaded inference after the repository\n  dependencies are installed.\n- State directory checks for `model.safetensors.index.json` and safetensors\n  weight-map keys used by `build_model`.\n- The notebook's model setup pattern: tokenizer, quant configs, `OffloadConfig`,\n  `build_model`, streamer, sampling parameters, and cached generation.\n- Memory and speed trade-offs for `offload_per_layer`.\n\n## What to route elsewhere\n\n- Use the root [..\u002F..\u002Freferences\u002Finstallation-and-runtime.md](..\u002F..\u002Freferences\u002Finstallation-and-runtime.md)\n  for source-only installation, `PYTHONPATH`, and package import checks.\n- Use [..\u002Fquantization-kernels\u002FSKILL.md](..\u002Fquantization-kernels\u002FSKILL.md) for\n  HQQ layer internals, packing functions, and Triton kernel errors.\n- Use [..\u002Fexpert-cache\u002FSKILL.md](..\u002Fexpert-cache\u002FSKILL.md) for LRU expert-cache,\n  storage, and SparseMoeWrapper internals.\n\n## Minimal decision flow\n\n1. Confirm the user has a CUDA-capable PyTorch runtime for actual inference.\n   CPU-only checks can validate imports and config math, but not the advertised\n   offloaded generation path.\n2. Confirm the quantized state path is local and contains a Hugging Face\n   safetensors index. Do not start a large network download unless the user\n   explicitly asks.\n3. Compute `offload_per_layer` and derived cache sizes from the model config.\n4. Build the quantization configs exactly before calling `build_model`.\n5. Keep generation code explicit about `attention_mask` and `past_key_values`\n   if continuing multi-turn chat.\n\n## Verification expectations\n\nSafe checks are `--help` on the bundled scripts and dry-run config rendering.\nFull Mixtral generation requires large external model artifacts and a CUDA GPU;\nclassify it as an approved native run only when the user accepts downloads,\nmemory use, and runtime cost.\n","# Inference Workflow\n\nUse this sub-skill when the task is to run, script, adapt, or troubleshoot the\nMixtral offloading demo workflow: building an offloaded `MixtralForCausalLM`\nfrom quantized safetensors, choosing offload sizes, and generating chat text.\n\n## Read this when\n\n- The user asks to turn the demo notebook into a Python script.\n- The task mentions `OffloadConfig`, `QuantConfig`, `build_model`,\n  `offload_per_layer`, or a quantized Mixtral state directory.\n- The user needs to balance GPU VRAM, system RAM, and generation speed.\n- A local run fails before or during model construction or generation.\n\n## Route map\n\n1. Read [references\u002Fworkflow.md](references\u002Fworkflow.md) for the full\n   notebook-to-script recipe, validation gates, and generation-loop outline.\n2. Read [references\u002Fconfiguration.md](references\u002Fconfiguration.md) when the task\n   is about offload sizing, HQQ quantization settings, or memory trade-offs.\n3. Use [scripts\u002Fcreate_offload_config.py](scripts\u002Fcreate_offload_config.py) to\n   compute `main_size`, `offload_size`, and `buffer_size` from layer\u002Fexpert\n   counts without importing the repository.\n4. Use [scripts\u002Frender_generation_skeleton.py](scripts\u002Frender_generation_skeleton.py)\n   to render a safe starter script that shows where to call `build_model`\n   without downloading weights or executing the model.\n5. Read [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for\n   state-path, CUDA, memory, HQQ config, tokenizer, and generation-cache errors.\n\n## What this sub-skill owns\n\n- Runtime prerequisites specific to offloaded inference after the repository\n  dependencies are installed.\n- State directory checks for `model.safetensors.index.json` and safetensors\n  weight-map keys used by `build_model`.\n- The notebook's model setup pattern: tokenizer, quant configs, `OffloadConfig`,\n  `build_model`, streamer, sampling parameters, and cached generation.\n- Memory and speed trade-offs for `offload_per_layer`.\n\n## What to route elsewhere\n\n- Use the root [..\u002F..\u002Freferences\u002Finstallation-and-runtime.md](..\u002F..\u002Freferences\u002Finstallation-and-runtime.md)\n  for source-only installation, `PYTHONPATH`, and package import checks.\n- Use [..\u002Fquantization-kernels\u002FSKILL.md](..\u002Fquantization-kernels\u002FSKILL.md) for\n  HQQ layer internals, packing functions, and Triton kernel errors.\n- Use [..\u002Fexpert-cache\u002FSKILL.md](..\u002Fexpert-cache\u002FSKILL.md) for LRU expert-cache,\n  storage, and SparseMoeWrapper internals.\n\n## Minimal decision flow\n\n1. Confirm the user has a CUDA-capable PyTorch runtime for actual inference.\n   CPU-only checks can validate imports and config math, but not the advertised\n   offloaded generation path.\n2. Confirm the quantized state path is local and contains a Hugging Face\n   safetensors index. Do not start a large network download unless the user\n   explicitly asks.\n3. Compute `offload_per_layer` and derived cache sizes from the model config.\n4. Build the quantization configs exactly before calling `build_model`.\n5. Keep generation code explicit about `attention_mask` and `past_key_values`\n   if continuing multi-turn chat.\n\n## Verification expectations\n\nSafe checks are `--help` on the bundled scripts and dry-run config rendering.\nFull Mixtral generation requires large external model artifacts and a CUDA GPU;\nclassify it as an approved native run only when the user accepts downloads,\nmemory use, and runtime cost.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow",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 inference-workflow",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill inference-workflow",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill inference-workflow","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002FSKILL.md",[46,52,56,60,66],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fconfiguration.md","REFERENCE","参考文档",2642,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002Freferences\u002Fconfiguration.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",2682,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fworkflow.md",4640,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002Freferences\u002Fworkflow.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fcreate_offload_config.py","SCRIPT","脚本",2348,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002Fscripts\u002Fcreate_offload_config.py",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Frender_generation_skeleton.py",4046,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmixtral-offloading\u002Fsub-skills\u002Finference-workflow\u002Fscripts\u002Frender_generation_skeleton.py",[71,79,86,93,100,107],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"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":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"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":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"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":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"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":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"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":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"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",[],[]]