[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78WG3MDCSKEEG12VJYVT8VR":3,"skill-same:01M78WG3MDCSKEEG12VJYVT8VR":66},{"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},"01M78WG3MDCSKEEG12VJYVT8VR","dreamerv2",null,"Use DreamerV2 2.2.0 for TensorFlow world-model reinforcement-learning training, custom Gym integration, typed experiment configuration, replay\u002Fcheckpoint management, and JSONL\u002FTensorBoard evaluation.","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\u002Fdreamerv2\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: dreamerv2\ndescription: \"Use DreamerV2 2.2.0 for TensorFlow world-model\n  reinforcement-learning training, custom Gym integration, typed experiment\n  configuration, replay\u002Fcheckpoint management, and JSONL\u002FTensorBoard\n  evaluation.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# DreamerV2\n\nUse this skill for the `danijar\u002Fdreamerv2` 2.2.0 package: discrete world-model\nreinforcement learning with TensorFlow, native Atari\u002FDM Control\u002FCrafter runners,\ncustom legacy Gym environments, typed configuration, replay\u002Fcheckpoints, and\nexperiment plots. Treat this as a TensorFlow 2.6-era package, not a current\nGymnasium or modern-Keras package.\n\n## Route the task\n\n- **Launch, bound, resume, or diagnose a run:** read\n  [training](sub-skills\u002Ftraining\u002FSKILL.md).\n- **Adapt an environment or validate observations\u002Factions:** read\n  [environments](sub-skills\u002Fenvironments\u002FSKILL.md).\n- **Compose presets, flags, schedules, or serialized configs:** read\n  [configuration](sub-skills\u002Fconfiguration\u002FSKILL.md).\n- **Inspect metrics, TensorBoard artifacts, or compare runs:** read\n  [evaluation](sub-skills\u002Fevaluation\u002FSKILL.md).\n\nRead only the route needed for the request, then follow its linked references.\nThe sub-skills are one operating graph: a training request commonly visits\nconfiguration first, environments for the data contract, and evaluation after\nartifacts exist.\n\n## Install a compatible runtime\n\nThe verified package facts are distribution `dreamerv2==2.2.0`, import modules\n`dreamerv2` and `dreamerv2.api`, and console entry point `dreamerv2`. This source\nrelease uses `tensorflow.keras.mixed_precision.experimental` and the legacy Gym\nAPI. Prefer a fresh Python 3.9 environment with a coherent era stack:\n\n```sh\npython -m pip install \\\n  'tensorflow==2.6.0' 'tensorflow-probability==0.14.1' \\\n  'gym==0.23.1' 'ruamel.yaml\u003C0.18' \\\n  'numpy==1.19.5' 'pillow\u003C10' 'pandas\u003C2' 'matplotlib\u003C3.8'\npython -m pip install --no-deps 'dreamerv2==2.2.0'\n```\n\nAdd only the suites you will use: `atari_py` plus imported Atari ROMs for\nAtari, a compatible `dm_control`\u002FMuJoCo\u002FEGL setup for DMC, `crafter` for\nCrafter, and the environment's own package for custom Gym. Do not download\nROMs, MuJoCo assets, or credentials as an unattended smoke check. Read\n[troubleshooting](references\u002Ftroubleshooting.md) before mixing newer TensorFlow,\nTensorFlow Probability, Gym\u002FGymnasium, or NumPy versions.\n\nCheck the installation without starting training:\n\n```sh\npython -m pip check\npython - \u003C\u003C'PY'\nimport inspect\nimport tensorflow as tf\nimport dreamerv2.api as dv2\nprint('TensorFlow', tf.__version__)\nprint('GPUs', tf.config.list_physical_devices('GPU'))\nprint('api.train', inspect.signature(dv2.train))\nprint('default task', dv2.defaults.task)\nPY\npython -m dreamerv2.train --help\n```\n\nThe module help route is verified. Do **not** use the installed `dreamerv2`\nlauncher in this release: `train.py` derives `configs.yaml` from\n`sys.argv[0]`, so the launcher may search beside its bin script and fail with a\nmissing-config error. Use `python -m dreamerv2.train` instead. The plotting\nroute is available through the evaluation sub-skill's bundled adapter.\n\n## Operating rules\n\n- Native `python -m dreamerv2.train` asserts that TensorFlow sees a GPU; a CPU\n  import or `debug` preset does not make native training CPU-safe.\n- Use a new writable `logdir` per run. The effective `config.yaml`, replay\n  episodes, `metrics.jsonl`, TensorBoard events, and `variables.pkl` are the\n  observable outputs; a checkpoint alone is not a successful experiment.\n- Validate one environment reset\u002Fstep and its action\u002Fobservation contract\n  before allocating training. Keep `--envs_parallel none` for this snapshot\n  unless the asynchronous branch has been patched and tested.\n- Make `debug` the last preset, set a bounded `--steps`, and explicitly choose\n  `--precision 32` when the legacy mixed-precision path is not verified.\n- Preserve exact package\u002Fruntime versions and effective config with results.\n  For refresh decisions, read [repository provenance](references\u002Frepo-provenance.md).\n\n## Scope and limits\n\nThis graph distills public package behavior and safe operational recipes. It\ndoes not run benchmark-scale training, fetch external assets, import the skill\ninto the managed repo library, or claim that a modern unpinned dependency set\nis compatible. Use [cross-cutting troubleshooting](references\u002Ftroubleshooting.md)\nfor install\u002Fimport, backend, dependency, and source-version failures before\nescalating to the focused route.\n","# DreamerV2\n\nUse this skill for the `danijar\u002Fdreamerv2` 2.2.0 package: discrete world-model\nreinforcement learning with TensorFlow, native Atari\u002FDM Control\u002FCrafter runners,\ncustom legacy Gym environments, typed configuration, replay\u002Fcheckpoints, and\nexperiment plots. Treat this as a TensorFlow 2.6-era package, not a current\nGymnasium or modern-Keras package.\n\n## Route the task\n\n- **Launch, bound, resume, or diagnose a run:** read\n  [training](sub-skills\u002Ftraining\u002FSKILL.md).\n- **Adapt an environment or validate observations\u002Factions:** read\n  [environments](sub-skills\u002Fenvironments\u002FSKILL.md).\n- **Compose presets, flags, schedules, or serialized configs:** read\n  [configuration](sub-skills\u002Fconfiguration\u002FSKILL.md).\n- **Inspect metrics, TensorBoard artifacts, or compare runs:** read\n  [evaluation](sub-skills\u002Fevaluation\u002FSKILL.md).\n\nRead only the route needed for the request, then follow its linked references.\nThe sub-skills are one operating graph: a training request commonly visits\nconfiguration first, environments for the data contract, and evaluation after\nartifacts exist.\n\n## Install a compatible runtime\n\nThe verified package facts are distribution `dreamerv2==2.2.0`, import modules\n`dreamerv2` and `dreamerv2.api`, and console entry point `dreamerv2`. This source\nrelease uses `tensorflow.keras.mixed_precision.experimental` and the legacy Gym\nAPI. Prefer a fresh Python 3.9 environment with a coherent era stack:\n\n```sh\npython -m pip install \\\n  'tensorflow==2.6.0' 'tensorflow-probability==0.14.1' \\\n  'gym==0.23.1' 'ruamel.yaml\u003C0.18' \\\n  'numpy==1.19.5' 'pillow\u003C10' 'pandas\u003C2' 'matplotlib\u003C3.8'\npython -m pip install --no-deps 'dreamerv2==2.2.0'\n```\n\nAdd only the suites you will use: `atari_py` plus imported Atari ROMs for\nAtari, a compatible `dm_control`\u002FMuJoCo\u002FEGL setup for DMC, `crafter` for\nCrafter, and the environment's own package for custom Gym. Do not download\nROMs, MuJoCo assets, or credentials as an unattended smoke check. Read\n[troubleshooting](references\u002Ftroubleshooting.md) before mixing newer TensorFlow,\nTensorFlow Probability, Gym\u002FGymnasium, or NumPy versions.\n\nCheck the installation without starting training:\n\n```sh\npython -m pip check\npython - \u003C\u003C'PY'\nimport inspect\nimport tensorflow as tf\nimport dreamerv2.api as dv2\nprint('TensorFlow', tf.__version__)\nprint('GPUs', tf.config.list_physical_devices('GPU'))\nprint('api.train', inspect.signature(dv2.train))\nprint('default task', dv2.defaults.task)\nPY\npython -m dreamerv2.train --help\n```\n\nThe module help route is verified. Do **not** use the installed `dreamerv2`\nlauncher in this release: `train.py` derives `configs.yaml` from\n`sys.argv[0]`, so the launcher may search beside its bin script and fail with a\nmissing-config error. Use `python -m dreamerv2.train` instead. The plotting\nroute is available through the evaluation sub-skill's bundled adapter.\n\n## Operating rules\n\n- Native `python -m dreamerv2.train` asserts that TensorFlow sees a GPU; a CPU\n  import or `debug` preset does not make native training CPU-safe.\n- Use a new writable `logdir` per run. The effective `config.yaml`, replay\n  episodes, `metrics.jsonl`, TensorBoard events, and `variables.pkl` are the\n  observable outputs; a checkpoint alone is not a successful experiment.\n- Validate one environment reset\u002Fstep and its action\u002Fobservation contract\n  before allocating training. Keep `--envs_parallel none` for this snapshot\n  unless the asynchronous branch has been patched and tested.\n- Make `debug` the last preset, set a bounded `--steps`, and explicitly choose\n  `--precision 32` when the legacy mixed-precision path is not verified.\n- Preserve exact package\u002Fruntime versions and effective config with results.\n  For refresh decisions, read [repository provenance](references\u002Frepo-provenance.md).\n\n## Scope and limits\n\nThis graph distills public package behavior and safe operational recipes. It\ndoes not run benchmark-scale training, fetch external assets, import the skill\ninto the managed repo library, or claim that a modern unpinned dependency set\nis compatible. Use [cross-cutting troubleshooting](references\u002Ftroubleshooting.md)\nfor install\u002Fimport, backend, dependency, and source-version failures before\nescalating to the focused route.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fdreamerv2",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 dreamerv2",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill dreamerv2",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill dreamerv2","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdreamerv2\u002FSKILL.md",[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Frepo-provenance.md","REFERENCE","参考文档",1645,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdreamerv2\u002Freferences\u002Frepo-provenance.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Frepo-routing-metadata.json",436,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdreamerv2\u002Freferences\u002Frepo-routing-metadata.json",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",5000,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdreamerv2\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fcheck_install.py","SCRIPT","脚本",1384,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdreamerv2\u002Fscripts\u002Fcheck_install.py",[67,75,82,89,96,103],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"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":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"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":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"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":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"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",[],[]]