[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EWQSDECYXVG7MZC0DP0HJH":3,"skill-same:01M7EWQSDECYXVG7MZC0DP0HJH":66},{"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},"01M7EWQSDECYXVG7MZC0DP0HJH","ai-optimizer",null,"Route AI-Optimizer reinforcement-learning collection tasks across model-based RL, easy-MARL, offline RL, safe command builders, and repository limitations.","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\u002F05-技能与代理开发\u002Fai-optimizer\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: ai-optimizer\ndescription: \"Route AI-Optimizer reinforcement-learning collection tasks across\n  model-based RL, easy-MARL, offline RL, safe command builders, and repository\n  limitations.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# AI-Optimizer\n\nUse this repo skill when a task names AI-Optimizer or asks for help with its reinforcement-learning algorithm collection: model-based RL, easy-MARL multi-agent RL, offline RL, algorithm selection, command construction, dataset\u002Fschema checks, dependency caveats, or safe static verification before running experiments.\n\nAI-Optimizer is a research-code collection rather than one installable Python distribution. Treat each algorithm family as its own runtime surface with separate dependencies and simulator\u002Fdata requirements.\n\n## Route by task\n\n| User request mentions | Read |\n|---|---|\n| Dreamer, ED2-Dreamer, PlaNet, MuZero, Sampled MuZero, MBPO, ED2-MBPO, BMPO, CaDM, world models, learned dynamics, planning, model-based baselines | [model-based-rl](sub-skills\u002Fmodel-based-rl\u002FSKILL.md) |\n| MARL, easy-MARL, IDQN, VDN, QMIX, CommNet, IDDPG, MADDPG, IPPO, MAPPO, MAGYM, MPE, scenario names, multi-agent training commands | [multi-agent-rl](sub-skills\u002Fmulti-agent-rl\u002FSKILL.md) |\n| Offline RL, batch RL, D4RL, MDPDataset, d3rlpy-derived APIs, BCQ, BEAR, CQL, AWAC, REDQ, UWAC, ISPI, COMBO, MOPO, E2O, PEX, offline-to-online | [offline-rl](sub-skills\u002Foffline-rl\u002FSKILL.md) |\n| Unsure where a folder or algorithm belongs, or need a high-level map of checked-in and empty areas | [repository-map.md](references\u002Frepository-map.md) |\n| Import\u002Finstall\u002Fbackend\u002Fdata failures shared across algorithm families | [troubleshooting.md](references\u002Ftroubleshooting.md) |\n| Staleness, source commit, dirty state, or evidence-path audit | [repo-provenance.md](references\u002Frepo-provenance.md) |\n\n## Safe default workflow\n\n1. Identify the algorithm family and route to the matching sub-skill.\n2. Use bundled command builders or validators before running original training code:\n   - Model-based MuZero command builder: `sub-skills\u002Fmodel-based-rl\u002Fscripts\u002Fbuild_muzero_command.py`.\n   - Easy-MARL command builder: `sub-skills\u002Fmulti-agent-rl\u002Fscripts\u002Fbuild_easy_marl_command.py`.\n   - Offline RL command builders and dataset validator: `sub-skills\u002Foffline-rl\u002Fscripts\u002F`.\n3. Run [scripts\u002Fcheck_ai_optimizer_static.py](scripts\u002Fcheck_ai_optimizer_static.py) when checking a generated skill tree or a target checkout layout without launching training.\n4. Before executing any printed command, verify the target runtime environment for that algorithm family. Do not assume dependencies are shared across subdirectories.\n5. Treat full RL training, simulator execution, CUDA allocation, dataset downloads, and background experiment launches as deliberate heavy actions that need task-specific approval and resources.\n\n## Important constraints\n\n- The inspected checkout contains checked-in code for `modelbased-rl`, `multiagent-rl\u002Feasy-marl`, and `offline-rl-algorithms`.\n- Several submodule placeholders are empty in the inspected source snapshot: `cornerstone`, `self-supervised-rl`, `transfer-and-multi-task-reinforcement-learning`, and `multiagent-rl\u002Fcore`. Do not claim their code is available from this skill.\n- Many algorithm folders target old ML stacks: TensorFlow 1.x\u002F2.1\u002F2.2 GPU, Ray 0.6\u002F0.7, old Gym, MuJoCo 1.50, dm_control, D4RL, Waymo, MAGYM, or MPE. Keep those prerequisites explicit.\n- This skill preserves operating guidance and safe helpers. It does not verify benchmark scores or completed training runs.\n- Runtime links in this skill point to bundled skill files. Source-path names in references are identifiers for target AI-Optimizer checkouts, not links to this production checkout.\n\n## Minimal inspection checks\n\nFor a target AI-Optimizer checkout, prefer safe checks before training:\n\n```bash\npython scripts\u002Fcheck_ai_optimizer_static.py --source-root \u002Fpath\u002Fto\u002FAI-Optimizer\npython sub-skills\u002Fmulti-agent-rl\u002Fscripts\u002Fbuild_easy_marl_command.py --agent-name IDQN --env-name discrete_meeting\npython sub-skills\u002Fmodel-based-rl\u002Fscripts\u002Fbuild_muzero_command.py --env CartPole-v1 --case classic_control --opr train --no-cuda --force\npython sub-skills\u002Foffline-rl\u002Fscripts\u002Fbuild_offline_rl_command.py bcq --dataset halfcheetah-medium-v2 --seed 0 --omit-gpu\n```\n\nThe printed commands are recipes for a target checkout. The helpers do not install dependencies, start environments, download datasets, run training, or write model outputs.\n","# AI-Optimizer\n\nUse this repo skill when a task names AI-Optimizer or asks for help with its reinforcement-learning algorithm collection: model-based RL, easy-MARL multi-agent RL, offline RL, algorithm selection, command construction, dataset\u002Fschema checks, dependency caveats, or safe static verification before running experiments.\n\nAI-Optimizer is a research-code collection rather than one installable Python distribution. Treat each algorithm family as its own runtime surface with separate dependencies and simulator\u002Fdata requirements.\n\n## Route by task\n\n| User request mentions | Read |\n|---|---|\n| Dreamer, ED2-Dreamer, PlaNet, MuZero, Sampled MuZero, MBPO, ED2-MBPO, BMPO, CaDM, world models, learned dynamics, planning, model-based baselines | [model-based-rl](sub-skills\u002Fmodel-based-rl\u002FSKILL.md) |\n| MARL, easy-MARL, IDQN, VDN, QMIX, CommNet, IDDPG, MADDPG, IPPO, MAPPO, MAGYM, MPE, scenario names, multi-agent training commands | [multi-agent-rl](sub-skills\u002Fmulti-agent-rl\u002FSKILL.md) |\n| Offline RL, batch RL, D4RL, MDPDataset, d3rlpy-derived APIs, BCQ, BEAR, CQL, AWAC, REDQ, UWAC, ISPI, COMBO, MOPO, E2O, PEX, offline-to-online | [offline-rl](sub-skills\u002Foffline-rl\u002FSKILL.md) |\n| Unsure where a folder or algorithm belongs, or need a high-level map of checked-in and empty areas | [repository-map.md](references\u002Frepository-map.md) |\n| Import\u002Finstall\u002Fbackend\u002Fdata failures shared across algorithm families | [troubleshooting.md](references\u002Ftroubleshooting.md) |\n| Staleness, source commit, dirty state, or evidence-path audit | [repo-provenance.md](references\u002Frepo-provenance.md) |\n\n## Safe default workflow\n\n1. Identify the algorithm family and route to the matching sub-skill.\n2. Use bundled command builders or validators before running original training code:\n   - Model-based MuZero command builder: `sub-skills\u002Fmodel-based-rl\u002Fscripts\u002Fbuild_muzero_command.py`.\n   - Easy-MARL command builder: `sub-skills\u002Fmulti-agent-rl\u002Fscripts\u002Fbuild_easy_marl_command.py`.\n   - Offline RL command builders and dataset validator: `sub-skills\u002Foffline-rl\u002Fscripts\u002F`.\n3. Run [scripts\u002Fcheck_ai_optimizer_static.py](scripts\u002Fcheck_ai_optimizer_static.py) when checking a generated skill tree or a target checkout layout without launching training.\n4. Before executing any printed command, verify the target runtime environment for that algorithm family. Do not assume dependencies are shared across subdirectories.\n5. Treat full RL training, simulator execution, CUDA allocation, dataset downloads, and background experiment launches as deliberate heavy actions that need task-specific approval and resources.\n\n## Important constraints\n\n- The inspected checkout contains checked-in code for `modelbased-rl`, `multiagent-rl\u002Feasy-marl`, and `offline-rl-algorithms`.\n- Several submodule placeholders are empty in the inspected source snapshot: `cornerstone`, `self-supervised-rl`, `transfer-and-multi-task-reinforcement-learning`, and `multiagent-rl\u002Fcore`. Do not claim their code is available from this skill.\n- Many algorithm folders target old ML stacks: TensorFlow 1.x\u002F2.1\u002F2.2 GPU, Ray 0.6\u002F0.7, old Gym, MuJoCo 1.50, dm_control, D4RL, Waymo, MAGYM, or MPE. Keep those prerequisites explicit.\n- This skill preserves operating guidance and safe helpers. It does not verify benchmark scores or completed training runs.\n- Runtime links in this skill point to bundled skill files. Source-path names in references are identifiers for target AI-Optimizer checkouts, not links to this production checkout.\n\n## Minimal inspection checks\n\nFor a target AI-Optimizer checkout, prefer safe checks before training:\n\n```bash\npython scripts\u002Fcheck_ai_optimizer_static.py --source-root \u002Fpath\u002Fto\u002FAI-Optimizer\npython sub-skills\u002Fmulti-agent-rl\u002Fscripts\u002Fbuild_easy_marl_command.py --agent-name IDQN --env-name discrete_meeting\npython sub-skills\u002Fmodel-based-rl\u002Fscripts\u002Fbuild_muzero_command.py --env CartPole-v1 --case classic_control --opr train --no-cuda --force\npython sub-skills\u002Foffline-rl\u002Fscripts\u002Fbuild_offline_rl_command.py bcq --dataset halfcheetah-medium-v2 --seed 0 --omit-gpu\n```\n\nThe printed commands are recipes for a target checkout. The helpers do not install dependencies, start environments, download datasets, run training, or write model outputs.\n",{"name":5,"license":15,"metadata":24,"description":7,"disable-model-invocation":26},{"disco-role":25},"operating",true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","05-技能与代理开发\u002Fai-optimizer",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 ai-optimizer",{"tool":41,"command":42},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill ai-optimizer",{"tool":44,"command":45},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill ai-optimizer","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fai-optimizer\u002FSKILL.md",[48,54,58,62],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"references\u002Frepo-provenance.md","REFERENCE","参考文档",3638,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fai-optimizer\u002Freferences\u002Frepo-provenance.md",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"references\u002Frepo-routing-metadata.json",532,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fai-optimizer\u002Freferences\u002Frepo-routing-metadata.json",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"references\u002Frepository-map.md",3028,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fai-optimizer\u002Freferences\u002Frepository-map.md",{"path":63,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":64,"sourceHtmlUrl":65},"references\u002Ftroubleshooting.md",3377,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fai-optimizer\u002Freferences\u002Ftroubleshooting.md",[67,75,80,87,94,101],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":74},"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":76,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":74},"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":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":74},"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":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":74},"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":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":74},"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":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":74},"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",[],[]]