[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78S5KPJ5V3RNFGGGK71AC3V":3,"skill-same:01M78S5KPJ5V3RNFGGGK71AC3V":78},{"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},"01M78S5KPJ5V3RNFGGGK71AC3V","adversarial-robustness-toolbox",null,"Use Adversarial Robustness Toolbox (ART) for estimator wrappers, adversarial attacks, defences, poisoning\u002Fprivacy\u002Fextraction, metrics, and certification 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\u002Fadversarial-robustness-toolbox\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: adversarial-robustness-toolbox\ndescription: \"Use Adversarial Robustness Toolbox (ART) for estimator wrappers,\n  adversarial attacks, defences, poisoning\u002Fprivacy\u002Fextraction, metrics, and\n  certification workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Adversarial Robustness Toolbox (ART)\n\nUse this repo skill when a task names `adversarial-robustness-toolbox`, `art`, ART estimators, adversarial examples, robustness defences, poisoning\u002Fbackdoor attacks, privacy inference attacks, model extraction, robustness metrics, or certification workflows.\n\nART is a Python library for wrapping ML models as estimators, attacking and defending them, and evaluating robustness. This skill covers the core CPU-capable Python workflows verified for ART 1.20.1; heavy speech, object-detection, tracking, malware, GAN\u002Fgeneration, experimental, notebook, and container-only workflows are treated as out of selected runtime scope unless a future refresh expands them.\n\n## Start here\n\n1. If ART is not installed or imports fail, go to [`sub-skills\u002Fsetup-and-backends\u002F`](sub-skills\u002Fsetup-and-backends\u002FSKILL.md).\n2. For a quick installed-environment check, run the bundled diagnostic from this skill directory:\n\n   ```bash\n   python scripts\u002Finspect_art_install.py --json\n   ```\n\n3. If a model is not yet wrapped as an ART estimator, use [`sub-skills\u002Festimators-and-models\u002F`](sub-skills\u002Festimators-and-models\u002FSKILL.md) before choosing attacks or metrics.\n4. Choose the workflow owner from the route table below, then read that sub-skill's references and bundled scripts.\n5. Before deciding whether this skill matches a newer checkout, read [`references\u002Frepo-provenance.md`](references\u002Frepo-provenance.md).\n\n## Route map\n\n| User request | Load this sub-skill | Why |\n|---|---|---|\n| Install ART, verify `import art`, choose CPU\u002FGPU packages, resolve optional dependency or backend errors | [`setup-and-backends`](sub-skills\u002Fsetup-and-backends\u002FSKILL.md) | Owns package groups, backend matrix, and import diagnostics. |\n| Wrap sklearn, PyTorch, TensorFlow\u002FKeras, boosted-tree, GPy, black-box, or regression models for ART | [`estimators-and-models`](sub-skills\u002Festimators-and-models\u002FSKILL.md) | Owns estimator constructor contracts, labels, shapes, `clip_values`, gradients, and black-box limitations. |\n| Generate evasion adversarial examples, choose FGSM\u002FPGD\u002FAutoAttack\u002FHopSkipJump\u002Fpatch attacks, add preprocessors, or plan adversarial training | [`evasion-and-preprocessing`](sub-skills\u002Fevasion-and-preprocessing\u002FSKILL.md) | Owns attack-family compatibility, perturbation budgets, preprocessing defences, and trainer recipes. |\n| Plan poisoning\u002Fbackdoor, membership\u002Fattribute inference, model inversion\u002Freconstruction, model extraction\u002Fstealing, or detector\u002Fmitigation workflows | [`poisoning-inference-extraction`](sub-skills\u002Fpoisoning-inference-extraction\u002FSKILL.md) | Owns poisoning, privacy inference, extraction, Neural Cleanse, Activation Defence, STRIP, and query\u002Fdata split guidance. |\n| Compute robustness\u002Fprivacy metrics, run SecurityCurve or GREAT score, use SummaryWriter routing, run tree verification, or choose certification wrappers | [`evaluation-and-certification`](sub-skills\u002Fevaluation-and-certification\u002FSKILL.md) | Owns ART metrics, evaluation objects, gradient checks, tree verification, randomized smoothing, DeepZ, and interval\u002FIBP routes. |\n\n## Shared references\n\n- [`references\u002Fbackend-matrix.md`](references\u002Fbackend-matrix.md) — install groups, optional backend families, and CPU\u002FGPU notes used across sub-skills.\n- [`references\u002Fworkflow-overview.md`](references\u002Fworkflow-overview.md) — end-to-end workflow order, input\u002Flabel conventions, and cross-sub-skill handoffs.\n- [`references\u002Fexamples-and-tests-map.md`](references\u002Fexamples-and-tests-map.md) — bundled helper map and native-candidate categories, without requiring source examples at runtime.\n- [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md) — cross-cutting install\u002Fimport, data shape, optional dependency, and workflow-routing failures.\n- [`references\u002Frepo-routing-metadata.json`](references\u002Frepo-routing-metadata.json) — structured router placement consumed during managed repo-skill import.\n\n## Public install check\n\nFor a normal user environment, start with the base package and only add backend libraries needed by the selected workflow:\n\n```bash\npython -m pip install adversarial-robustness-toolbox\npython -c \"import art; print(art.__version__)\"\n```\n\nFor PyTorch CPU workflows, be explicit in ART wrappers that accept a device parameter:\n\n```python\nclassifier = PyTorchClassifier(..., device_type=\"cpu\")\n```\n\nART's PyTorch wrappers often default to GPU-oriented device selection, so this explicit CPU choice avoids no-CUDA confusion.\n\n## Guardrails\n\n- Do not run original repository examples, notebooks, tests, or CI scripts as runtime instructions for users. Use the bundled scripts and distilled references in this skill tree.\n- Do not claim a black-box estimator supports white-box attacks requiring `loss_gradient`; route to compatible attacks or rebuild a gradient-enabled estimator.\n- Do not treat successful attack generation as proof of robustness. Use the metrics\u002Fcertification sub-skill for evaluation, stronger attacks, and gradient-obfuscation checks.\n- Do not import or update the managed repo-skill library from this runtime skill. Import is a separate verified creation workflow step.\n","# Adversarial Robustness Toolbox (ART)\n\nUse this repo skill when a task names `adversarial-robustness-toolbox`, `art`, ART estimators, adversarial examples, robustness defences, poisoning\u002Fbackdoor attacks, privacy inference attacks, model extraction, robustness metrics, or certification workflows.\n\nART is a Python library for wrapping ML models as estimators, attacking and defending them, and evaluating robustness. This skill covers the core CPU-capable Python workflows verified for ART 1.20.1; heavy speech, object-detection, tracking, malware, GAN\u002Fgeneration, experimental, notebook, and container-only workflows are treated as out of selected runtime scope unless a future refresh expands them.\n\n## Start here\n\n1. If ART is not installed or imports fail, go to [`sub-skills\u002Fsetup-and-backends\u002F`](sub-skills\u002Fsetup-and-backends\u002FSKILL.md).\n2. For a quick installed-environment check, run the bundled diagnostic from this skill directory:\n\n   ```bash\n   python scripts\u002Finspect_art_install.py --json\n   ```\n\n3. If a model is not yet wrapped as an ART estimator, use [`sub-skills\u002Festimators-and-models\u002F`](sub-skills\u002Festimators-and-models\u002FSKILL.md) before choosing attacks or metrics.\n4. Choose the workflow owner from the route table below, then read that sub-skill's references and bundled scripts.\n5. Before deciding whether this skill matches a newer checkout, read [`references\u002Frepo-provenance.md`](references\u002Frepo-provenance.md).\n\n## Route map\n\n| User request | Load this sub-skill | Why |\n|---|---|---|\n| Install ART, verify `import art`, choose CPU\u002FGPU packages, resolve optional dependency or backend errors | [`setup-and-backends`](sub-skills\u002Fsetup-and-backends\u002FSKILL.md) | Owns package groups, backend matrix, and import diagnostics. |\n| Wrap sklearn, PyTorch, TensorFlow\u002FKeras, boosted-tree, GPy, black-box, or regression models for ART | [`estimators-and-models`](sub-skills\u002Festimators-and-models\u002FSKILL.md) | Owns estimator constructor contracts, labels, shapes, `clip_values`, gradients, and black-box limitations. |\n| Generate evasion adversarial examples, choose FGSM\u002FPGD\u002FAutoAttack\u002FHopSkipJump\u002Fpatch attacks, add preprocessors, or plan adversarial training | [`evasion-and-preprocessing`](sub-skills\u002Fevasion-and-preprocessing\u002FSKILL.md) | Owns attack-family compatibility, perturbation budgets, preprocessing defences, and trainer recipes. |\n| Plan poisoning\u002Fbackdoor, membership\u002Fattribute inference, model inversion\u002Freconstruction, model extraction\u002Fstealing, or detector\u002Fmitigation workflows | [`poisoning-inference-extraction`](sub-skills\u002Fpoisoning-inference-extraction\u002FSKILL.md) | Owns poisoning, privacy inference, extraction, Neural Cleanse, Activation Defence, STRIP, and query\u002Fdata split guidance. |\n| Compute robustness\u002Fprivacy metrics, run SecurityCurve or GREAT score, use SummaryWriter routing, run tree verification, or choose certification wrappers | [`evaluation-and-certification`](sub-skills\u002Fevaluation-and-certification\u002FSKILL.md) | Owns ART metrics, evaluation objects, gradient checks, tree verification, randomized smoothing, DeepZ, and interval\u002FIBP routes. |\n\n## Shared references\n\n- [`references\u002Fbackend-matrix.md`](references\u002Fbackend-matrix.md) — install groups, optional backend families, and CPU\u002FGPU notes used across sub-skills.\n- [`references\u002Fworkflow-overview.md`](references\u002Fworkflow-overview.md) — end-to-end workflow order, input\u002Flabel conventions, and cross-sub-skill handoffs.\n- [`references\u002Fexamples-and-tests-map.md`](references\u002Fexamples-and-tests-map.md) — bundled helper map and native-candidate categories, without requiring source examples at runtime.\n- [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md) — cross-cutting install\u002Fimport, data shape, optional dependency, and workflow-routing failures.\n- [`references\u002Frepo-routing-metadata.json`](references\u002Frepo-routing-metadata.json) — structured router placement consumed during managed repo-skill import.\n\n## Public install check\n\nFor a normal user environment, start with the base package and only add backend libraries needed by the selected workflow:\n\n```bash\npython -m pip install adversarial-robustness-toolbox\npython -c \"import art; print(art.__version__)\"\n```\n\nFor PyTorch CPU workflows, be explicit in ART wrappers that accept a device parameter:\n\n```python\nclassifier = PyTorchClassifier(..., device_type=\"cpu\")\n```\n\nART's PyTorch wrappers often default to GPU-oriented device selection, so this explicit CPU choice avoids no-CUDA confusion.\n\n## Guardrails\n\n- Do not run original repository examples, notebooks, tests, or CI scripts as runtime instructions for users. Use the bundled scripts and distilled references in this skill tree.\n- Do not claim a black-box estimator supports white-box attacks requiring `loss_gradient`; route to compatible attacks or rebuild a gradient-enabled estimator.\n- Do not treat successful attack generation as proof of robustness. Use the metrics\u002Fcertification sub-skill for evaluation, stronger attacks, and gradient-obfuscation checks.\n- Do not import or update the managed repo-skill library from this runtime skill. Import is a separate verified creation workflow step.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox",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 adversarial-robustness-toolbox",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill adversarial-robustness-toolbox",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill adversarial-robustness-toolbox","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002FSKILL.md",[46,52,56,60,64,68,72],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fbackend-matrix.md","REFERENCE","参考文档",4635,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Freferences\u002Fbackend-matrix.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fexamples-and-tests-map.md",4361,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Freferences\u002Fexamples-and-tests-map.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Frepo-provenance.md",1988,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Freferences\u002Frepo-provenance.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Frepo-routing-metadata.json",453,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Freferences\u002Frepo-routing-metadata.json",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Ftroubleshooting.md",4162,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Freferences\u002Ftroubleshooting.md",{"path":69,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Fworkflow-overview.md",3847,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Freferences\u002Fworkflow-overview.md",{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"scripts\u002Finspect_art_install.py","SCRIPT","脚本",918,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fadversarial-robustness-toolbox\u002Fscripts\u002Finspect_art_install.py",[79,87,94,101,108,115],{"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":83,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"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":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":83,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"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":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":83,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"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":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":83,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"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":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":83,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"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":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":83,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":121,"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",[],[]]