[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7FBYK1GW80GEN2KMPEX19AB":3,"skill-same:01M7FBYK1GW80GEN2KMPEX19AB":67},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M7FBYK1GW80GEN2KMPEX19AB","simulation-environments",null,"Install and operate the legacy Rex-Gym Gym\u002FPyBullet quadruped environments, including task, signal, terrain, asset, control-loop, and safe-smoke-test decisions.","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: simulation-environments\ndescription: \"Install and operate the legacy Rex-Gym Gym\u002FPyBullet quadruped\n  environments, including task, signal, terrain, asset, control-loop, and\n  safe-smoke-test decisions.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# Rex-Gym simulation environments\n\nUse this skill when a task mentions Rex-Gym, a Rex quadruped simulation, a\nPyBullet Gym task, locomotion environments, terrains, action spaces, or the\n`rex-gym` CLI's `env`, `terrain`, or `mark` arguments. It covers the packaged\nlegacy environments and their Gym-facing behavior, not policy training.\n\n## Route before acting\n\n- For inverse-kinematics math, gait trajectories, and motor-model details, use\n  [locomotion-modeling](..\u002Flocomotion-modeling\u002FSKILL.md).\n- For PPO graphs, training configuration, checkpoints, or policy playback, use\n  [training-policy](..\u002Ftraining-policy\u002FSKILL.md).\n- Use the [task matrix](references\u002Ftask-matrix.md) to choose a task and signal,\n  the [API reference](references\u002Fapi-reference.md) for constructor and Gym\n  contracts, and [terrain and assets](references\u002Fterrain-and-assets.md) before\n  selecting a non-plane terrain.\n- Run the bundled [headless smoke script](scripts\u002Fsmoke_environment.py) for a\n  bounded environment check. It never trains, downloads, or opens a GUI unless\n  `--render` is explicitly supplied.\n- Diagnose failures with [troubleshooting](references\u002Ftroubleshooting.md).\n\n## Install the package\n\nThe public package name is `rex_gym` (the Python import root is `rex_gym`):\n\n```bash\npython -m pip install rex_gym\n```\n\nThis is a legacy Python 3.7-era project. Prefer an isolated Python 3.7\nruntime when reproducing the pinned stack (`gym` 0.17-era APIs, NumPy 1.17,\nand PyBullet 2.8-era behavior). Environment-only imports use Gym, NumPy, and\nPyBullet; the PPO CLI surface additionally imports TensorFlow 1.x and\nTensorFlow Probability 0.8, so do not install or debug that optional surface\nfor a simulation smoke unless training or playback is requested.\n\n## Select a task and signal\n\nThe CLI names are `poses`, `gallop`, `walk`, `turn`, and `standup`. The source\nmaps their defaults to `ik`, `ik`, `ik`, `ol`, and `ol`, respectively. Override\nwith exactly one of `--inverse-kinematics` (`-ik`) or `--open-loop` (`-ol`).\nThe task-specific action dimensions and bounds are in the\n[task matrix](references\u002Ftask-matrix.md); subclasses replace the base class's\n12\u002F18-motor action space with compact signal-feedback actions.\n\nThe direct Python class imports are:\n\n```python\nfrom rex_gym.envs.gym.poses_env import RexPosesEnv\nfrom rex_gym.envs.gym.gallop_env import RexReactiveEnv\nfrom rex_gym.envs.gym.walk_env import RexWalkEnv\nfrom rex_gym.envs.gym.turn_env import RexTurnEnv\nfrom rex_gym.envs.gym.standup_env import RexStandupEnv\n```\n\nFor direct construction, pass an explicit terrain id, including for the\nordinary plane. This is required by the current Rex initialization lookup:\n\n```python\nfrom rex_gym.envs.gym.walk_env import RexWalkEnv\n\nenv = RexWalkEnv(\n    render=False, terrain_type=\"plane\", terrain_id=\"plane\",\n    signal_type=\"ik\", target_position=2.0, backwards=False, mark=\"base\")\ntry:\n    observation = env.reset()\n    observation, reward, done, info = env.step([0.0, 0.0])\nfinally:\n    env.close()\n```\n\n`RexGymEnv(render=False, terrain_type=\"plane\", terrain_id=\"plane\")` is the\nlow-level environment. It exposes direct desired-motor-angle control; the\nfive task classes above add compact pose or gait signal controllers. Do not\nreimplement those controllers here; route their math to\n[locomotion-modeling](..\u002Flocomotion-modeling\u002FSKILL.md).\n\n## CLI mapping\n\nThe CLI converts `--terrain NAME` into both `terrain_type` and `terrain_id`,\nconverts `--mark NAME` into `mark`, and converts repeated `--arg KEY VALUE`\npairs into constructor keyword arguments; repeated `--flag KEY BOOL` pairs are\nmerged into the same argument dictionary for boolean constructor keywords.\nRelevant examples:\n\n```bash\nrex-gym --help\nrex-gym train --env walk --log-dir .\u002Flogs --terrain plane --mark base \\\n  --inverse-kinematics --arg target_position 2\nrex-gym train --env turn --log-dir .\u002Flogs --terrain hills --mark base \\\n  --open-loop --arg init_orient 0.5 --arg target_orient 3.0\nrex-gym policy --env poses --terrain plane --mark base --inverse-kinematics\n```\n\n`--env` accepts the names in the legacy mapper, including a `go` entry, but\nthis package's Gym environment files\u002Fregister table do not provide a usable\n`RexGoEnv`; use the five tasks above. `train` and `policy` load TensorFlow\u002FPPO\nsurfaces and may render or run indefinitely; they are not smoke commands.\nFollow [training-policy](..\u002Ftraining-policy\u002FSKILL.md) for those workflows.\n\n## Operate the legacy Gym loop\n\nThe API is the old four-return Gym API, not the newer five-return API:\n\n```python\nobservation = env.reset()\nfor _ in range(20):\n    action = env.action_space.sample()  # only when its declared Box is valid\n    observation, reward, done, info = env.step(action)\n    if done:\n        break\nenv.close()\n```\n\nUse the action dimensions and task bounds from the matrix, and prefer a\nzero\u002Fknown-safe action for smoke tests. `info` contains an `action` key with\nthe transformed motor command. `done` can mean a fall, a reached target, or\nan out-of-trajectory state, depending on the task. `close()` calls the Rex\ntermination hook; always use `try\u002Ffinally` so Bullet objects can be released.\n\nBase observations contain motor angles, velocities, torques, and a base\nquaternion: length `3*N + 4` (`40` for `base`, `58` for `arm`). Task\nobservations are compact: poses, walk, turn, and standup expose roll, pitch,\nand their rates (length 4); gallop adds 12 motor angles (length 16 for base).\nThe exact mark-dependent extensions and live quirks are recorded in\n[api-reference](references\u002Fapi-reference.md).\n\n## GUI and headless decisions\n\nUse `render=False` (or the smoke script default) on servers, CI, and displayless\nshells. `render=True` opens a PyBullet GUI during construction and therefore\nneeds a usable display; it is not a headless renderer. GUI IK modes add sliders\nfor base position\u002Forientation and, for gait tasks, gait parameters. Poses also\nsupports direct base\u002Forientation slider control. `render(mode=\"rgb_array\")`\nreturns an image; the legacy implementation returns an empty array for other\nmodes. Use GUI only for interactive inspection, not unattended validation.\n\n## Acceptance check\n\nA useful environment check must establish import, construction, reset, action\nshape, one or more bounded steps, legacy return structure, `info[\"action\"]`,\nand cleanup. Run:\n\n```bash\npython scripts\u002Fsmoke_environment.py --task walk --signal ik --terrain random \\\n  --mark base --target-position 1.0 --direction forward --steps 2\npython scripts\u002Fsmoke_environment.py --help\n```\n\nThe helper accepts `--target-position` for `walk`\u002F`gallop` and\n`--direction forward|backward|random` for `walk`; it reports the constructor\nkeywords it used. For walk open loop, use `--signal ol` and the helper's\nexplicit zero action of shape `(8,)`. If a particular packaged terrain fails\non reset, preserve the reported error and switch to `plane` for a repeatable\nCPU smoke rather than retrying indefinitely.\n\nThe script emits one JSON object and reports missing dependencies, unsupported\nasset\u002Fterrain combinations, invalid values, or display errors without hiding\nthe exception class. See [troubleshooting](references\u002Ftroubleshooting.md) for\nsource-backed recovery steps.\n","# Rex-Gym simulation environments\n\nUse this skill when a task mentions Rex-Gym, a Rex quadruped simulation, a\nPyBullet Gym task, locomotion environments, terrains, action spaces, or the\n`rex-gym` CLI's `env`, `terrain`, or `mark` arguments. It covers the packaged\nlegacy environments and their Gym-facing behavior, not policy training.\n\n## Route before acting\n\n- For inverse-kinematics math, gait trajectories, and motor-model details, use\n  [locomotion-modeling](..\u002Flocomotion-modeling\u002FSKILL.md).\n- For PPO graphs, training configuration, checkpoints, or policy playback, use\n  [training-policy](..\u002Ftraining-policy\u002FSKILL.md).\n- Use the [task matrix](references\u002Ftask-matrix.md) to choose a task and signal,\n  the [API reference](references\u002Fapi-reference.md) for constructor and Gym\n  contracts, and [terrain and assets](references\u002Fterrain-and-assets.md) before\n  selecting a non-plane terrain.\n- Run the bundled [headless smoke script](scripts\u002Fsmoke_environment.py) for a\n  bounded environment check. It never trains, downloads, or opens a GUI unless\n  `--render` is explicitly supplied.\n- Diagnose failures with [troubleshooting](references\u002Ftroubleshooting.md).\n\n## Install the package\n\nThe public package name is `rex_gym` (the Python import root is `rex_gym`):\n\n```bash\npython -m pip install rex_gym\n```\n\nThis is a legacy Python 3.7-era project. Prefer an isolated Python 3.7\nruntime when reproducing the pinned stack (`gym` 0.17-era APIs, NumPy 1.17,\nand PyBullet 2.8-era behavior). Environment-only imports use Gym, NumPy, and\nPyBullet; the PPO CLI surface additionally imports TensorFlow 1.x and\nTensorFlow Probability 0.8, so do not install or debug that optional surface\nfor a simulation smoke unless training or playback is requested.\n\n## Select a task and signal\n\nThe CLI names are `poses`, `gallop`, `walk`, `turn`, and `standup`. The source\nmaps their defaults to `ik`, `ik`, `ik`, `ol`, and `ol`, respectively. Override\nwith exactly one of `--inverse-kinematics` (`-ik`) or `--open-loop` (`-ol`).\nThe task-specific action dimensions and bounds are in the\n[task matrix](references\u002Ftask-matrix.md); subclasses replace the base class's\n12\u002F18-motor action space with compact signal-feedback actions.\n\nThe direct Python class imports are:\n\n```python\nfrom rex_gym.envs.gym.poses_env import RexPosesEnv\nfrom rex_gym.envs.gym.gallop_env import RexReactiveEnv\nfrom rex_gym.envs.gym.walk_env import RexWalkEnv\nfrom rex_gym.envs.gym.turn_env import RexTurnEnv\nfrom rex_gym.envs.gym.standup_env import RexStandupEnv\n```\n\nFor direct construction, pass an explicit terrain id, including for the\nordinary plane. This is required by the current Rex initialization lookup:\n\n```python\nfrom rex_gym.envs.gym.walk_env import RexWalkEnv\n\nenv = RexWalkEnv(\n    render=False, terrain_type=\"plane\", terrain_id=\"plane\",\n    signal_type=\"ik\", target_position=2.0, backwards=False, mark=\"base\")\ntry:\n    observation = env.reset()\n    observation, reward, done, info = env.step([0.0, 0.0])\nfinally:\n    env.close()\n```\n\n`RexGymEnv(render=False, terrain_type=\"plane\", terrain_id=\"plane\")` is the\nlow-level environment. It exposes direct desired-motor-angle control; the\nfive task classes above add compact pose or gait signal controllers. Do not\nreimplement those controllers here; route their math to\n[locomotion-modeling](..\u002Flocomotion-modeling\u002FSKILL.md).\n\n## CLI mapping\n\nThe CLI converts `--terrain NAME` into both `terrain_type` and `terrain_id`,\nconverts `--mark NAME` into `mark`, and converts repeated `--arg KEY VALUE`\npairs into constructor keyword arguments; repeated `--flag KEY BOOL` pairs are\nmerged into the same argument dictionary for boolean constructor keywords.\nRelevant examples:\n\n```bash\nrex-gym --help\nrex-gym train --env walk --log-dir .\u002Flogs --terrain plane --mark base \\\n  --inverse-kinematics --arg target_position 2\nrex-gym train --env turn --log-dir .\u002Flogs --terrain hills --mark base \\\n  --open-loop --arg init_orient 0.5 --arg target_orient 3.0\nrex-gym policy --env poses --terrain plane --mark base --inverse-kinematics\n```\n\n`--env` accepts the names in the legacy mapper, including a `go` entry, but\nthis package's Gym environment files\u002Fregister table do not provide a usable\n`RexGoEnv`; use the five tasks above. `train` and `policy` load TensorFlow\u002FPPO\nsurfaces and may render or run indefinitely; they are not smoke commands.\nFollow [training-policy](..\u002Ftraining-policy\u002FSKILL.md) for those workflows.\n\n## Operate the legacy Gym loop\n\nThe API is the old four-return Gym API, not the newer five-return API:\n\n```python\nobservation = env.reset()\nfor _ in range(20):\n    action = env.action_space.sample()  # only when its declared Box is valid\n    observation, reward, done, info = env.step(action)\n    if done:\n        break\nenv.close()\n```\n\nUse the action dimensions and task bounds from the matrix, and prefer a\nzero\u002Fknown-safe action for smoke tests. `info` contains an `action` key with\nthe transformed motor command. `done` can mean a fall, a reached target, or\nan out-of-trajectory state, depending on the task. `close()` calls the Rex\ntermination hook; always use `try\u002Ffinally` so Bullet objects can be released.\n\nBase observations contain motor angles, velocities, torques, and a base\nquaternion: length `3*N + 4` (`40` for `base`, `58` for `arm`). Task\nobservations are compact: poses, walk, turn, and standup expose roll, pitch,\nand their rates (length 4); gallop adds 12 motor angles (length 16 for base).\nThe exact mark-dependent extensions and live quirks are recorded in\n[api-reference](references\u002Fapi-reference.md).\n\n## GUI and headless decisions\n\nUse `render=False` (or the smoke script default) on servers, CI, and displayless\nshells. `render=True` opens a PyBullet GUI during construction and therefore\nneeds a usable display; it is not a headless renderer. GUI IK modes add sliders\nfor base position\u002Forientation and, for gait tasks, gait parameters. Poses also\nsupports direct base\u002Forientation slider control. `render(mode=\"rgb_array\")`\nreturns an image; the legacy implementation returns an empty array for other\nmodes. Use GUI only for interactive inspection, not unattended validation.\n\n## Acceptance check\n\nA useful environment check must establish import, construction, reset, action\nshape, one or more bounded steps, legacy return structure, `info[\"action\"]`,\nand cleanup. Run:\n\n```bash\npython scripts\u002Fsmoke_environment.py --task walk --signal ik --terrain random \\\n  --mark base --target-position 1.0 --direction forward --steps 2\npython scripts\u002Fsmoke_environment.py --help\n```\n\nThe helper accepts `--target-position` for `walk`\u002F`gallop` and\n`--direction forward|backward|random` for `walk`; it reports the constructor\nkeywords it used. For walk open loop, use `--signal ol` and the helper's\nexplicit zero action of shape `(8,)`. If a particular packaged terrain fails\non reset, preserve the reported error and switch to `plane` for a repeatable\nCPU smoke rather than retrying indefinitely.\n\nThe script emits one JSON object and reports missing dependencies, unsupported\nasset\u002Fterrain combinations, invalid values, or display errors without hiding\nthe exception class. See [troubleshooting](references\u002Ftroubleshooting.md) for\nsource-backed recovery steps.\n",{"name":5,"license":24,"metadata":25,"description":7,"disable-model-invocation":27},"Apache 2.0",{"disco-role":26},"operating",true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments",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":37,"prompt":47,"zipAvailable":27},[38,41,44],{"tool":39,"command":40},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill simulation-environments",{"tool":42,"command":43},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill simulation-environments",{"tool":45,"command":46},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill simulation-environments","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments\u002FSKILL.md",[49,55,59,63],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fapi-reference.md","REFERENCE","参考文档",5832,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments\u002Freferences\u002Fapi-reference.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Ftask-matrix.md",5374,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments\u002Freferences\u002Ftask-matrix.md",{"path":60,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fterrain-and-assets.md",4228,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments\u002Freferences\u002Fterrain-and-assets.md",{"path":64,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":65,"sourceHtmlUrl":66},"references\u002Ftroubleshooting.md",6111,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F06-前端与设计\u002Frex-gym\u002Fsub-skills\u002Fsimulation-environments\u002Freferences\u002Ftroubleshooting.md",[68,76,81,88,95,102],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"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.","MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],{"skillNo":77,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"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":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"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":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"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":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":19},"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":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":19},"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",[],[]]