[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EDZFHQJMSCZ3CS8ZXS09W9":3,"skill-same:01M7EDZFHQJMSCZ3CS8ZXS09W9":62},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":33,"updatedAt":34,"install":35,"resources":47},"01M7EDZFHQJMSCZ3CS8ZXS09W9","evasion-and-preprocessing",null,"Use ART evasion attacks, preprocessing defences, and adversarial training recipes for image\u002Ftabular robustness workflows.","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fadversarial-robustness-toolbox\u002Fsub-skills\u002Fevasion-and-preprocessing\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: evasion-and-preprocessing\ndescription: \"Use ART evasion attacks, preprocessing defences, and adversarial\n  training recipes for image\u002Ftabular robustness workflows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Evasion and preprocessing\n\nUse this sub-skill when the task is to generate evasion adversarial examples, select a white-box\u002Fblack-box\u002Fphysical image attack, add preprocessing defences, or plan adversarial training with ART estimators.\n\n## Route here\n\n- Craft FGSM\u002FFGM, PGD\u002FBIM\u002FMIM\u002FAutoPGD\u002FAutoAttack, Carlini, DeepFool, ElasticNet, JSMA, universal perturbation, virtual adversarial, HopSkipJump, Boundary, ZOO, SimBA, Square, Pixel\u002FThreshold, SignOPT, DecisionTree, spatial, feature-adversary, or image-patch attacks.\n- Add image\u002Ftabular preprocessing defences such as standardisation, spatial smoothing, feature squeezing, JPEG compression, Gaussian augmentation, label smoothing, Mixup, CutMix, Cutout, thermometer encoding, total variation minimization, or PixelDefend.\n- Plan adversarial training with `AdversarialTrainer`, `AdversarialTrainerMadryPGD`, or TRADES-style trainers.\n\n## Route away\n\n- Model wrapping, `clip_values`, `input_shape`, backend device setup, and estimator construction details belong to `..\u002Festimators-and-models\u002F` or `..\u002Fsetup-and-backends\u002F`.\n- Poisoning, backdoors, membership\u002Fattribute inference, model inversion, or model extraction belong to `..\u002Fpoisoning-inference-extraction\u002F`.\n- Robustness metrics, certification, tree verification, security curves, and gradient checks after attack generation belong to `..\u002Fevaluation-and-certification\u002F`.\n- Speech, object-detection-heavy, malware, and audio-heavy evasion workflows are outside this selected runtime scope unless a future refresh adds their required backends. Recognize their ART class names, but do not promise runnable coverage here.\n\n## Operating sequence\n\n1. Confirm the estimator capability: loss gradients for PGD\u002FFGM\u002FBIM\u002FMIM\u002FAutoPGD, class gradients for Carlini\u002FDeepFool\u002FJSMA-style attacks, prediction-only classifiers for hard-label black-box attacks, or neural-network\u002Fimage-specific support for patch\u002Fspatial attacks.\n2. Match all perturbation budgets to the estimator input scale and `clip_values`; for normalized `[0, 1]` images use budgets such as `8\u002F255`, not `8`.\n3. Check shape conventions before running an attack: PyTorch image estimators usually use `channels_first=True` and `NCHW`; many NumPy image preprocessors default to `channels_first=False` and `NHWC`.\n4. For targeted attacks, pass target labels as `y` to `generate`; use one-hot labels unless the wrapped estimator workflow explicitly uses class-index labels.\n5. Add preprocessing defences either directly as callable preprocessors or through estimator `preprocessing_defences`; do not treat preprocessing alone as proof of robustness.\n6. For adversarial training, start from a bounded PGD\u002FFGM attack, verify labels and clipping, then choose generic, Madry PGD, or TRADES training according to the estimator backend.\n\n## References\n\n- Attack family\u002FAPI chooser: [references\u002Fattack-api-reference.md](references\u002Fattack-api-reference.md)\n- Preprocessing and adversarial training recipes: [references\u002Fpreprocessing-and-training.md](references\u002Fpreprocessing-and-training.md)\n- Evasion\u002Fpreprocessing troubleshooting: [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md)\n\n## Bundled checks\n\n- Tiny CPU PyTorch PGD\u002FFGM smoke: [scripts\u002Fsmoke_evasion_pytorch.py](scripts\u002Fsmoke_evasion_pytorch.py)\n- Tiny NumPy standardisation\u002Fspatial smoothing smoke: [scripts\u002Fsmoke_preprocessor_numpy.py](scripts\u002Fsmoke_preprocessor_numpy.py)\n\nRun bundled scripts with `--help` first; they use synthetic data and do not download datasets.\n","# Evasion and preprocessing\n\nUse this sub-skill when the task is to generate evasion adversarial examples, select a white-box\u002Fblack-box\u002Fphysical image attack, add preprocessing defences, or plan adversarial training with ART estimators.\n\n## Route here\n\n- Craft FGSM\u002FFGM, PGD\u002FBIM\u002FMIM\u002FAutoPGD\u002FAutoAttack, Carlini, DeepFool, ElasticNet, JSMA, universal perturbation, virtual adversarial, HopSkipJump, Boundary, ZOO, SimBA, Square, Pixel\u002FThreshold, SignOPT, DecisionTree, spatial, feature-adversary, or image-patch attacks.\n- Add image\u002Ftabular preprocessing defences such as standardisation, spatial smoothing, feature squeezing, JPEG compression, Gaussian augmentation, label smoothing, Mixup, CutMix, Cutout, thermometer encoding, total variation minimization, or PixelDefend.\n- Plan adversarial training with `AdversarialTrainer`, `AdversarialTrainerMadryPGD`, or TRADES-style trainers.\n\n## Route away\n\n- Model wrapping, `clip_values`, `input_shape`, backend device setup, and estimator construction details belong to `..\u002Festimators-and-models\u002F` or `..\u002Fsetup-and-backends\u002F`.\n- Poisoning, backdoors, membership\u002Fattribute inference, model inversion, or model extraction belong to `..\u002Fpoisoning-inference-extraction\u002F`.\n- Robustness metrics, certification, tree verification, security curves, and gradient checks after attack generation belong to `..\u002Fevaluation-and-certification\u002F`.\n- Speech, object-detection-heavy, malware, and audio-heavy evasion workflows are outside this selected runtime scope unless a future refresh adds their required backends. Recognize their ART class names, but do not promise runnable coverage here.\n\n## Operating sequence\n\n1. Confirm the estimator capability: loss gradients for PGD\u002FFGM\u002FBIM\u002FMIM\u002FAutoPGD, class gradients for Carlini\u002FDeepFool\u002FJSMA-style attacks, prediction-only classifiers for hard-label black-box attacks, or neural-network\u002Fimage-specific support for patch\u002Fspatial attacks.\n2. Match all perturbation budgets to the estimator input scale and `clip_values`; for normalized `[0, 1]` images use budgets such as `8\u002F255`, not `8`.\n3. Check shape conventions before running an attack: PyTorch image estimators usually use `channels_first=True` and `NCHW`; many NumPy image preprocessors default to `channels_first=False` and `NHWC`.\n4. For targeted attacks, pass target labels as `y` to `generate`; use one-hot labels unless the wrapped estimator workflow explicitly uses class-index labels.\n5. Add preprocessing defences either directly as callable preprocessors or through estimator `preprocessing_defences`; do not treat preprocessing alone as proof of robustness.\n6. For adversarial training, start from a bounded PGD\u002FFGM attack, verify labels and clipping, then choose generic, Madry PGD, or TRADES training according to the estimator backend.\n\n## References\n\n- Attack family\u002FAPI chooser: [references\u002Fattack-api-reference.md](references\u002Fattack-api-reference.md)\n- Preprocessing and adversarial training recipes: [references\u002Fpreprocessing-and-training.md](references\u002Fpreprocessing-and-training.md)\n- Evasion\u002Fpreprocessing troubleshooting: [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md)\n\n## Bundled checks\n\n- Tiny CPU PyTorch PGD\u002FFGM smoke: [scripts\u002Fsmoke_evasion_pytorch.py](scripts\u002Fsmoke_evasion_pytorch.py)\n- Tiny NumPy standardisation\u002Fspatial smoothing smoke: [scripts\u002Fsmoke_preprocessor_numpy.py](scripts\u002Fsmoke_preprocessor_numpy.py)\n\nRun bundled scripts with `--help` first; they use synthetic data and do not download datasets.\n",{"name":5,"license":15,"metadata":24,"description":7,"disable-model-invocation":26},{"disco-role":25},"operating",true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","04-安全渗透\u002Fadversarial-robustness-toolbox\u002Fsub-skills\u002Fevasion-and-preprocessing",3,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts","https:\u002F\u002Fgithub.com\u002Flza6","个人vx：Tf00798\r\n个人✈：@TFGY999","2026-10-08 14:29:22",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":26},[37,40,43],{"tool":38,"command":39},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill evasion-and-preprocessing",{"tool":41,"command":42},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill evasion-and-preprocessing",{"tool":44,"command":45},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill evasion-and-preprocessing","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fadversarial-robustness-toolbox\u002Fsub-skills\u002Fevasion-and-preprocessing\u002FSKILL.md",[48,54,58],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"references\u002Fattack-api-reference.md","REFERENCE","参考文档",10991,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fadversarial-robustness-toolbox\u002Fsub-skills\u002Fevasion-and-preprocessing\u002Freferences\u002Fattack-api-reference.md",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"references\u002Fpreprocessing-and-training.md",9880,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fadversarial-robustness-toolbox\u002Fsub-skills\u002Fevasion-and-preprocessing\u002Freferences\u002Fpreprocessing-and-training.md",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"references\u002Ftroubleshooting.md",7201,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fadversarial-robustness-toolbox\u002Fsub-skills\u002Fevasion-and-preprocessing\u002Freferences\u002Ftroubleshooting.md",[63,71,76,83,90,97],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":70},"01M7FFG485KY83TD6HSC4YPCMC","obsidian-clipper-template-creator","Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],"2026-11-16 10:05:34",{"skillNo":72,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":70},"01M7FFG0T6AZ0STEJPSB3CX84K","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator-antigravity-awesome-skills-main\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":70},"01M7FFFY0QEM14SKWPDYGMXM2P","python-api","Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP\u002FmAP semantics, and source-style import diagnostics.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Fpython-api\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":70},"01M7FFFVZQZ2GJ0C0TGP7JHD20","file-evaluation","Folder-based PASCAL VOC AP\u002FmAP evaluation from ground-truth and detection text files.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Ffile-evaluation\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":70},"01M7FFFSFNYSE77SJ6PE1CK2DA","object-detection-metrics","Guides legacy Object-Detection-Metrics PASCAL VOC AP\u002FmAP evaluation from detection text folders and direct Python API classes.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":70},"01M7FFFP0GCK1RVXBCPQ8APBMT","nx-workspace-patterns","Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fnx-workspace-patterns\u002FSKILL.md",[],[]]