[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78T4JEB9745S5SEB2PBEA48":3,"skill-same:01M78T4JEB9745S5SEB2PBEA48":82},{"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},"01M78T4JEB9745S5SEB2PBEA48","background-matting-v2",null,"Routes BackgroundMattingV2 tasks for background matting inference, export, and training 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\u002Fbackground-matting-v2\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: background-matting-v2\ndescription: \"Routes BackgroundMattingV2 tasks for background matting inference,\n  export, and training workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# BackgroundMattingV2\n\nBackgroundMattingV2 is a computer-vision repo for high-resolution background\nmatting with a source-image \u002F background-image pair. Use this root skill as a\nrouter, not as a full manual.\n\n## Start here\n\n- Read `references\u002Frepo-provenance.md` when you need to check whether this skill\n  still matches the repository checkout.\n- Run `scripts\u002Fcheck_env.py` first when you only need a quick import and tiny\n  forward smoke.\n- Read `references\u002Fworkflows.md` for the high-level route map and the decision\n  points between inference, export, and training.\n- Read `references\u002Fapi-reference.md` for verified model and dataset signatures.\n- Read `references\u002Fdata-formats.md` before touching `data_path.py` or paired\n  foreground\u002Falpha\u002Fbackground directories.\n- Read `references\u002Fbackend-compatibility.md` before choosing PyTorch,\n  TorchScript, or ONNX paths.\n- Read `references\u002Ftroubleshooting.md` for cross-cutting install, import, and\n  backend issues.\n\n## Install and inspect\n\nThis repo is source-tree based rather than a packaged wheel. For inspection,\ncreate an isolated Python environment, install the runtime stack used by the\nrepo workflows, and then import the source modules from a checkout of this repo.\nThe verified inspection stack used for this skill was Python 3.11 with:\n\n- `torch` + matching `torchvision`\n- `kornia`\n- `opencv-python`\n- `onnx`\n- `onnxruntime`\n- `tensorboard`\n- `tqdm`\n\nA quick smoke is:\n\n```bash\npython scripts\u002Fcheck_env.py --repo-root \u003Crepo-checkout> --device cuda\n```\n\nUse `--device cpu` when you only need importability and tiny forward coverage. Add `onnx` when you want the ONNX smoke helper to validate export support.\n\n## Route map\n\n### Inference and demo\nUse `sub-skills\u002Finference-and-demo\u002F` when the task is about:\n\n- `inference_images.py`\n- `inference_video.py`\n- `inference_webcam.py`\n- `inference_speed_test.py`\n- choosing model type, backbone, refine mode, device, or output types\n- understanding source\u002Fbackground pairing and alignment behavior\n\nRead `sub-skills\u002Finference-and-demo\u002FSKILL.md` for the trigger terms and linked\nworkflow helpers.\n\n### Export and backend compatibility\nUse `sub-skills\u002Fexport-and-backends\u002F` when the task is about:\n\n- `export_torchscript.py`\n- `export_onnx.py`\n- TorchScript attribute hoisting\n- ONNX patch crop\u002Freplace compatibility choices\n- validating export\u002Fruntime combinations\n\nRead `sub-skills\u002Fexport-and-backends\u002FSKILL.md` when you need conversion steps or\nbackend troubleshooting.\n\n### Training and data setup\nUse `sub-skills\u002Ftraining\u002F` when the task is about:\n\n- `train_base.py`\n- `train_refine.py`\n- `data_path.py`\n- paired foreground\u002Falpha\u002Fbackground directory layout\n- training checkpoints, logs, and benchmark evaluation\n- CUDA\u002FDDP assumptions and dataset-name selection\n\nRead `sub-skills\u002Ftraining\u002FSKILL.md` before configuring data paths or starting a\ntraining run.\n\n## Public surface summary\n\nThe public source-root modules you are expected to know are `model`, `dataset`,\n`inference_utils`, and `data_path`. The primary classes are `MattingBase` and\n`MattingRefine`. The main inference CLIs work with source\u002Fbackground image or\nvideo pairs and can optionally apply homographic alignment.\n\n## What not to do\n\n- Do not send future agents back to the original repo docs or scripts when a\n  bundled reference or script exists here.\n- Do not assume training or webcam workflows are safe to run without the right\n  hardware, data, and display devices.\n- Do not treat CPU importability as proof of CUDA readiness.\n\n## Local entry points\n\n- `scripts\u002Fcheck_env.py`\n- `sub-skills\u002Finference-and-demo\u002Fscripts\u002Fsmoke_forward.py`\n- `sub-skills\u002Fexport-and-backends\u002Fscripts\u002Fcheck_export_support.py`\n- `sub-skills\u002Ftraining\u002Fscripts\u002Fcheck_data_layout.py`\n","# BackgroundMattingV2\n\nBackgroundMattingV2 is a computer-vision repo for high-resolution background\nmatting with a source-image \u002F background-image pair. Use this root skill as a\nrouter, not as a full manual.\n\n## Start here\n\n- Read `references\u002Frepo-provenance.md` when you need to check whether this skill\n  still matches the repository checkout.\n- Run `scripts\u002Fcheck_env.py` first when you only need a quick import and tiny\n  forward smoke.\n- Read `references\u002Fworkflows.md` for the high-level route map and the decision\n  points between inference, export, and training.\n- Read `references\u002Fapi-reference.md` for verified model and dataset signatures.\n- Read `references\u002Fdata-formats.md` before touching `data_path.py` or paired\n  foreground\u002Falpha\u002Fbackground directories.\n- Read `references\u002Fbackend-compatibility.md` before choosing PyTorch,\n  TorchScript, or ONNX paths.\n- Read `references\u002Ftroubleshooting.md` for cross-cutting install, import, and\n  backend issues.\n\n## Install and inspect\n\nThis repo is source-tree based rather than a packaged wheel. For inspection,\ncreate an isolated Python environment, install the runtime stack used by the\nrepo workflows, and then import the source modules from a checkout of this repo.\nThe verified inspection stack used for this skill was Python 3.11 with:\n\n- `torch` + matching `torchvision`\n- `kornia`\n- `opencv-python`\n- `onnx`\n- `onnxruntime`\n- `tensorboard`\n- `tqdm`\n\nA quick smoke is:\n\n```bash\npython scripts\u002Fcheck_env.py --repo-root \u003Crepo-checkout> --device cuda\n```\n\nUse `--device cpu` when you only need importability and tiny forward coverage. Add `onnx` when you want the ONNX smoke helper to validate export support.\n\n## Route map\n\n### Inference and demo\nUse `sub-skills\u002Finference-and-demo\u002F` when the task is about:\n\n- `inference_images.py`\n- `inference_video.py`\n- `inference_webcam.py`\n- `inference_speed_test.py`\n- choosing model type, backbone, refine mode, device, or output types\n- understanding source\u002Fbackground pairing and alignment behavior\n\nRead `sub-skills\u002Finference-and-demo\u002FSKILL.md` for the trigger terms and linked\nworkflow helpers.\n\n### Export and backend compatibility\nUse `sub-skills\u002Fexport-and-backends\u002F` when the task is about:\n\n- `export_torchscript.py`\n- `export_onnx.py`\n- TorchScript attribute hoisting\n- ONNX patch crop\u002Freplace compatibility choices\n- validating export\u002Fruntime combinations\n\nRead `sub-skills\u002Fexport-and-backends\u002FSKILL.md` when you need conversion steps or\nbackend troubleshooting.\n\n### Training and data setup\nUse `sub-skills\u002Ftraining\u002F` when the task is about:\n\n- `train_base.py`\n- `train_refine.py`\n- `data_path.py`\n- paired foreground\u002Falpha\u002Fbackground directory layout\n- training checkpoints, logs, and benchmark evaluation\n- CUDA\u002FDDP assumptions and dataset-name selection\n\nRead `sub-skills\u002Ftraining\u002FSKILL.md` before configuring data paths or starting a\ntraining run.\n\n## Public surface summary\n\nThe public source-root modules you are expected to know are `model`, `dataset`,\n`inference_utils`, and `data_path`. The primary classes are `MattingBase` and\n`MattingRefine`. The main inference CLIs work with source\u002Fbackground image or\nvideo pairs and can optionally apply homographic alignment.\n\n## What not to do\n\n- Do not send future agents back to the original repo docs or scripts when a\n  bundled reference or script exists here.\n- Do not assume training or webcam workflows are safe to run without the right\n  hardware, data, and display devices.\n- Do not treat CPU importability as proof of CUDA readiness.\n\n## Local entry points\n\n- `scripts\u002Fcheck_env.py`\n- `sub-skills\u002Finference-and-demo\u002Fscripts\u002Fsmoke_forward.py`\n- `sub-skills\u002Fexport-and-backends\u002Fscripts\u002Fcheck_export_support.py`\n- `sub-skills\u002Ftraining\u002Fscripts\u002Fcheck_data_layout.py`\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2",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 background-matting-v2",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill background-matting-v2",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill background-matting-v2","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002FSKILL.md",[46,52,56,60,64,68,72,76],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",3187,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fbackend-compatibility.md",1875,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Fbackend-compatibility.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fdata-formats.md",1550,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Fdata-formats.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Frepo-provenance.md",1480,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Frepo-provenance.md",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Frepo-routing-metadata.json",343,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Frepo-routing-metadata.json",{"path":69,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Ftroubleshooting.md",3405,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Ftroubleshooting.md",{"path":73,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fworkflows.md",2466,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Freferences\u002Fworkflows.md",{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"scripts\u002Fcheck_env.py","SCRIPT","脚本",3390,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fbackground-matting-v2\u002Fscripts\u002Fcheck_env.py",[83,91,98,105,112,119],{"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":87,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"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":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"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":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"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":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"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":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"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":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":87,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"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",[],[]]