[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7964CHXW4NTRECJJYMDHTYN":3,"skill-same:01M7964CHXW4NTRECJJYMDHTYN":75},{"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},"01M7964CHXW4NTRECJJYMDHTYN","tensorpack",null,"Use Tensorpack, a TF1-compatible TensorFlow training interface with DataFlow, trainers, callbacks, prediction, checkpoints, export, and example workflow guidance.","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\u002Ftensorpack\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: tensorpack\ndescription: \"Use Tensorpack, a TF1-compatible TensorFlow training interface\n  with DataFlow, trainers, callbacks, prediction, checkpoints, export, and\n  example workflow guidance.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# Tensorpack repo skill\n\nUse this repo skill when a task names Tensorpack or clearly needs its graph-mode\nTensorFlow training interface, pure-Python DataFlow pipeline, callbacks,\ncheckpoint utilities, model export, or documented example workflow patterns.\nTensorpack is TF1-oriented: even under a TensorFlow 2 package, guide users toward\n`tensorpack.tfv1` \u002F `tf.compat.v1` graph code and explicit eager-mode handling.\n\n## Start here\n\n1. Read [`references\u002Frepo-overview.md`](references\u002Frepo-overview.md) for the\n   package topology, dependency tiers, TensorFlow compatibility, and public\n   module map.\n2. Read [`references\u002Frepo-provenance.md`](references\u002Frepo-provenance.md) before\n   deciding whether this skill is current for another checkout.\n3. Use [`scripts\u002Fcheck_tensorpack_env.py`](scripts\u002Fcheck_tensorpack_env.py) for a\n   safe import and optional-dependency diagnostic.\n4. If the task names a source example family or paper\u002Fmodel recipe, read\n   [`references\u002Fexamples-catalog.md`](references\u002Fexamples-catalog.md) and then\n   route to the owning sub-skill.\n5. If the user reports an install\u002Fimport\u002Fbackend\u002Fdata failure before you know\n   the owning workflow, read [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md).\n\n## Route by task\n\n| User intent or signal | Read |\n| --- | --- |\n| Data loading, augmentation, serializers, dataset loaders, `DataFlow`, `QueueInput`, input queue speed, `MultiProcessRunnerZMQ`, `LMDBSerializer`, `imgaug` | [`sub-skills\u002Fdataflow\u002FSKILL.md`](sub-skills\u002Fdataflow\u002FSKILL.md) |\n| Model definitions, `ModelDesc`, `TrainConfig`, trainers, callbacks, summaries, symbolic layers, multi-GPU\u002Fdistributed training, training example recipes | [`sub-skills\u002Ftraining\u002FSKILL.md`](sub-skills\u002Ftraining\u002FSKILL.md) |\n| `PredictConfig`, `OfflinePredictor`, `SmartInit`, checkpoint\u002F`.npz` inspection, Caffe conversion, SavedModel export, compact graph export, inference example patterns | [`sub-skills\u002Finference-export\u002FSKILL.md`](sub-skills\u002Finference-export\u002FSKILL.md) |\n| Broad example-family lookup across basics, ResNet\u002FImageNet, Faster\u002FMask R-CNN, GAN, RL, speech, NLP, Caffe, saliency\u002FCAM, Keras | [`references\u002Fexamples-catalog.md`](references\u002Fexamples-catalog.md) |\n| Unknown install\u002Fimport\u002Fruntime failure | [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md), then the nearest sub-skill troubleshooting guide |\n\n## Use this skill when\n\n- The user asks about Tensorpack APIs such as `DataFlow`, `ModelDesc`,\n  `TrainConfig`, `SimpleTrainer`, `InferenceRunner`, `OfflinePredictor`,\n  `SmartInit`, `ModelExporter`, `argscope`, `LinearWrap`, or Tensorpack model\n  layers.\n- The user needs to build or debug a graph-mode TensorFlow training loop while\n  keeping data loading, callbacks, summaries, validation, checkpointing, and\n  export consistent with Tensorpack conventions.\n- The user has Tensorpack checkpoint or model-zoo weights and needs variable\n  names, shape matching, `.npz` conversion, prediction, or export.\n- The task is to adapt a Tensorpack example pattern, while preserving dataset,\n  optional dependency, and CPU\u002FGPU verification limits.\n\n## Avoid this skill when\n\n- The request is ordinary PyTorch, JAX, modern `tf.keras`, or TensorFlow 2 eager\n  training with no Tensorpack dependency.\n- The user wants a full benchmark or paper-performance reproduction but has not\n  supplied the exact dataset, pretrained weights, hardware, and optional\n  dependencies; provide implementation guidance and state verification limits.\n- The task is general TensorFlow Serving, container orchestration, or MLOps\n  deployment beyond Tensorpack export artifact creation.\n- The request is only about maintaining this repository checkout rather than\n  using Tensorpack as a package; use a repository-maintenance skill if available.\n\n## Install and import guidance\n\nTensorpack's package dependencies include NumPy, six, termcolor, tabulate, tqdm,\nmsgpack\u002Fmsgpack-numpy, pyzmq, and psutil. TensorFlow is required for training,\nmodels, prediction, checkpoints, and export; DataFlow can be used as a pure\nPython data pipeline without full TensorFlow workflows. OpenCV is optional but\nimportant for image augmentation and most vision examples.\n\nPublic install patterns from the project evidence:\n\n```bash\npip install --upgrade git+https:\u002F\u002Fgithub.com\u002Ftensorpack\u002Ftensorpack.git\n# or pin an exact Tensorpack source\u002Frelease version for reproducible research\n```\n\nMinimal import check:\n\n```bash\npython - \u003C\u003C'PY'\nimport tensorpack\nprint(\"tensorpack\", tensorpack.__version__)\nfrom tensorpack.dataflow import FakeData, BatchData\nfrom tensorpack import tfv1 as tf\nprint(\"tf compat\", tf.__name__)\nPY\n```\n\nFor TF2 runtimes, disable eager execution before graph construction when the\nworkflow uses Tensorpack trainers or graph APIs:\n\n```python\nfrom tensorpack import tfv1 as tf\ntf.disable_eager_execution()\n```\n\n## Safe helper\n\nRun the bundled environment diagnostic from the root of this skill directory or\nby giving its full path:\n\n```bash\npython scripts\u002Fcheck_tensorpack_env.py --json\npython scripts\u002Fcheck_tensorpack_env.py --require-tf --require-cv2\n```\n\nIt reports installed Tensorpack\u002FTensorFlow versions, optional dependency status,\nand visible TensorFlow GPU devices. It does not download data, train, write\noutside normal stdout, or inspect original source files.\n\n## Verification status\n\nThis generated skill was constructed with a CPU inspection and smoke environment.\nCore package imports, DataFlow construction, TensorFlow graph construction, and\nselected helper parser checks were verified. CUDA\u002Fmulti-GPU, Horovod\u002FBytePS,\nCaffe, Atari\u002FGym, COCO, TIMIT, and large-dataset example workflows are documented\nas optional or data\u002Fbackend-dependent; do not claim they are verified unless the\nuser supplies and verifies the required stack.\n\n## Hard boundaries\n\n- Keep runtime answers self-contained. Do not require future agents to open or\n  run original Tensorpack source files, docs, examples, or scripts.\n- Link only to files inside this generated skill tree when giving reusable\n  instructions. Source evidence paths live in provenance and review artifacts,\n  not as runtime dependencies.\n- Do not reveal private environment prefixes, activation commands, local checkout\n  paths, or package installation locations in user-facing answers.\n- Respect the user's current request: this skill was generated for local use and\n  should not be imported unless a later user explicitly requests import.\n","# Tensorpack repo skill\n\nUse this repo skill when a task names Tensorpack or clearly needs its graph-mode\nTensorFlow training interface, pure-Python DataFlow pipeline, callbacks,\ncheckpoint utilities, model export, or documented example workflow patterns.\nTensorpack is TF1-oriented: even under a TensorFlow 2 package, guide users toward\n`tensorpack.tfv1` \u002F `tf.compat.v1` graph code and explicit eager-mode handling.\n\n## Start here\n\n1. Read [`references\u002Frepo-overview.md`](references\u002Frepo-overview.md) for the\n   package topology, dependency tiers, TensorFlow compatibility, and public\n   module map.\n2. Read [`references\u002Frepo-provenance.md`](references\u002Frepo-provenance.md) before\n   deciding whether this skill is current for another checkout.\n3. Use [`scripts\u002Fcheck_tensorpack_env.py`](scripts\u002Fcheck_tensorpack_env.py) for a\n   safe import and optional-dependency diagnostic.\n4. If the task names a source example family or paper\u002Fmodel recipe, read\n   [`references\u002Fexamples-catalog.md`](references\u002Fexamples-catalog.md) and then\n   route to the owning sub-skill.\n5. If the user reports an install\u002Fimport\u002Fbackend\u002Fdata failure before you know\n   the owning workflow, read [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md).\n\n## Route by task\n\n| User intent or signal | Read |\n| --- | --- |\n| Data loading, augmentation, serializers, dataset loaders, `DataFlow`, `QueueInput`, input queue speed, `MultiProcessRunnerZMQ`, `LMDBSerializer`, `imgaug` | [`sub-skills\u002Fdataflow\u002FSKILL.md`](sub-skills\u002Fdataflow\u002FSKILL.md) |\n| Model definitions, `ModelDesc`, `TrainConfig`, trainers, callbacks, summaries, symbolic layers, multi-GPU\u002Fdistributed training, training example recipes | [`sub-skills\u002Ftraining\u002FSKILL.md`](sub-skills\u002Ftraining\u002FSKILL.md) |\n| `PredictConfig`, `OfflinePredictor`, `SmartInit`, checkpoint\u002F`.npz` inspection, Caffe conversion, SavedModel export, compact graph export, inference example patterns | [`sub-skills\u002Finference-export\u002FSKILL.md`](sub-skills\u002Finference-export\u002FSKILL.md) |\n| Broad example-family lookup across basics, ResNet\u002FImageNet, Faster\u002FMask R-CNN, GAN, RL, speech, NLP, Caffe, saliency\u002FCAM, Keras | [`references\u002Fexamples-catalog.md`](references\u002Fexamples-catalog.md) |\n| Unknown install\u002Fimport\u002Fruntime failure | [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md), then the nearest sub-skill troubleshooting guide |\n\n## Use this skill when\n\n- The user asks about Tensorpack APIs such as `DataFlow`, `ModelDesc`,\n  `TrainConfig`, `SimpleTrainer`, `InferenceRunner`, `OfflinePredictor`,\n  `SmartInit`, `ModelExporter`, `argscope`, `LinearWrap`, or Tensorpack model\n  layers.\n- The user needs to build or debug a graph-mode TensorFlow training loop while\n  keeping data loading, callbacks, summaries, validation, checkpointing, and\n  export consistent with Tensorpack conventions.\n- The user has Tensorpack checkpoint or model-zoo weights and needs variable\n  names, shape matching, `.npz` conversion, prediction, or export.\n- The task is to adapt a Tensorpack example pattern, while preserving dataset,\n  optional dependency, and CPU\u002FGPU verification limits.\n\n## Avoid this skill when\n\n- The request is ordinary PyTorch, JAX, modern `tf.keras`, or TensorFlow 2 eager\n  training with no Tensorpack dependency.\n- The user wants a full benchmark or paper-performance reproduction but has not\n  supplied the exact dataset, pretrained weights, hardware, and optional\n  dependencies; provide implementation guidance and state verification limits.\n- The task is general TensorFlow Serving, container orchestration, or MLOps\n  deployment beyond Tensorpack export artifact creation.\n- The request is only about maintaining this repository checkout rather than\n  using Tensorpack as a package; use a repository-maintenance skill if available.\n\n## Install and import guidance\n\nTensorpack's package dependencies include NumPy, six, termcolor, tabulate, tqdm,\nmsgpack\u002Fmsgpack-numpy, pyzmq, and psutil. TensorFlow is required for training,\nmodels, prediction, checkpoints, and export; DataFlow can be used as a pure\nPython data pipeline without full TensorFlow workflows. OpenCV is optional but\nimportant for image augmentation and most vision examples.\n\nPublic install patterns from the project evidence:\n\n```bash\npip install --upgrade git+https:\u002F\u002Fgithub.com\u002Ftensorpack\u002Ftensorpack.git\n# or pin an exact Tensorpack source\u002Frelease version for reproducible research\n```\n\nMinimal import check:\n\n```bash\npython - \u003C\u003C'PY'\nimport tensorpack\nprint(\"tensorpack\", tensorpack.__version__)\nfrom tensorpack.dataflow import FakeData, BatchData\nfrom tensorpack import tfv1 as tf\nprint(\"tf compat\", tf.__name__)\nPY\n```\n\nFor TF2 runtimes, disable eager execution before graph construction when the\nworkflow uses Tensorpack trainers or graph APIs:\n\n```python\nfrom tensorpack import tfv1 as tf\ntf.disable_eager_execution()\n```\n\n## Safe helper\n\nRun the bundled environment diagnostic from the root of this skill directory or\nby giving its full path:\n\n```bash\npython scripts\u002Fcheck_tensorpack_env.py --json\npython scripts\u002Fcheck_tensorpack_env.py --require-tf --require-cv2\n```\n\nIt reports installed Tensorpack\u002FTensorFlow versions, optional dependency status,\nand visible TensorFlow GPU devices. It does not download data, train, write\noutside normal stdout, or inspect original source files.\n\n## Verification status\n\nThis generated skill was constructed with a CPU inspection and smoke environment.\nCore package imports, DataFlow construction, TensorFlow graph construction, and\nselected helper parser checks were verified. CUDA\u002Fmulti-GPU, Horovod\u002FBytePS,\nCaffe, Atari\u002FGym, COCO, TIMIT, and large-dataset example workflows are documented\nas optional or data\u002Fbackend-dependent; do not claim they are verified unless the\nuser supplies and verifies the required stack.\n\n## Hard boundaries\n\n- Keep runtime answers self-contained. Do not require future agents to open or\n  run original Tensorpack source files, docs, examples, or scripts.\n- Link only to files inside this generated skill tree when giving reusable\n  instructions. Source evidence paths live in provenance and review artifacts,\n  not as runtime dependencies.\n- Do not reveal private environment prefixes, activation commands, local checkout\n  paths, or package installation locations in user-facing answers.\n- Respect the user's current request: this skill was generated for local use and\n  should not be imported unless a later user explicitly requests import.\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\u002Ftensorpack",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 tensorpack",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill tensorpack",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill tensorpack","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002FSKILL.md",[47,53,57,61,65,69],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fexamples-catalog.md","REFERENCE","参考文档",7522,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002Freferences\u002Fexamples-catalog.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Frepo-overview.md",6591,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002Freferences\u002Frepo-overview.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Frepo-provenance.md",3180,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002Freferences\u002Frepo-provenance.md",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Frepo-routing-metadata.json",534,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002Freferences\u002Frepo-routing-metadata.json",{"path":66,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":67,"sourceHtmlUrl":68},"references\u002Ftroubleshooting.md",5644,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002Freferences\u002Ftroubleshooting.md",{"path":70,"resourceType":71,"resourceTypeDesc":72,"sizeBytes":73,"sourceHtmlUrl":74},"scripts\u002Fcheck_tensorpack_env.py","SCRIPT","脚本",4967,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftensorpack\u002Fscripts\u002Fcheck_tensorpack_env.py",[76,83,90,97,104,111],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"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.","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":14,"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":14,"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":14,"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":14,"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":14,"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",[],[]]