[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7946RD57YTP627ANNMYDY4Q":3,"skill-same:01M7946RD57YTP627ANNMYDY4Q":62},{"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},"01M7946RD57YTP627ANNMYDY4Q","grid-world-dp-and-control",null,"Use and adapt GridWorld dynamic programming, tabular control, Deep SARSA, and REINFORCE workflows distilled from the reinforcement-learning examples.","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\u002Freinforcement-learning\u002Fsub-skills\u002Fgrid-world-dp-and-control\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: grid-world-dp-and-control\ndescription: \"Use and adapt GridWorld dynamic programming, tabular control, Deep\n  SARSA, and REINFORCE workflows distilled from the reinforcement-learning\n  examples.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: MIT\n---\n\n# GridWorld DP and Control\n\nUse this sub-skill when the task is about the repository's 5x5 GridWorld workflows: policy iteration, value iteration, tabular SARSA, tabular Q-learning, Deep SARSA, or REINFORCE. It is self-contained; do not depend on the original checkout at runtime.\n\n## Route here for\n\n- Explaining, adapting, or debugging GridWorld dynamic programming (`1-policy_iteration.py`, `2-value_iteration.py`) as provenance workflow labels.\n- Explaining, adapting, or debugging tabular control (`3-sarsa.py`, `4-q_learning.py`) as provenance workflow labels.\n- Making tiny CPU checks for Deep SARSA and REINFORCE neural update logic (`5-deep_sarsa.py`, `6-reinforce.py`) as provenance workflow labels.\n- Diagnosing row\u002Fcolumn versus column\u002Frow coordinate mistakes, action-order mismatches, and render\u002Fheadless behavior.\n\n## Route away from\n\n- Gymnasium CartPole DQN\u002FA2C\u002FPPO: use the CartPole sub-skill.\n- Atari Breakout\u002FPong DQN\u002FPPO, preprocessing, ROMs, or W&B: use the standard Atari sub-skill.\n- Montezuma\u002FPitfall\u002FPrivateEye hard-exploration, RND, Go-Explore, or robustification: use the hard Atari sub-skill.\n\n## Start here\n\n1. Read [`references\u002Falgorithm-guide.md`](references\u002Falgorithm-guide.md) for the MDP layout, coordinate conventions, formulas, API-shaped examples, and safe adaptation patterns.\n2. Read [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md) when a grid policy points in the wrong direction, rendering blocks on a server, tabular values do not update, or Torch neural checks fail.\n3. Run the bundled smoke script when you need a self-contained, non-rendering API check:\n\n```bash\npython scripts\u002Fgrid_world_smoke.py --help\npython scripts\u002Fgrid_world_smoke.py\npython scripts\u002Fgrid_world_smoke.py --section dp\npython scripts\u002Fgrid_world_smoke.py --section neural --strict-torch\n```\n\nThe smoke script intentionally reimplements tiny fixtures and update steps. It does not import source repository files, open Pygame, train to convergence, or require a display.\n\n## Critical facts to preserve\n\n- DP workflows use state as `[row, col]`; drawing helpers usually need `(col, row)`, so swap before rendering arrows or markers.\n- Tabular `Env`-style workflows use state as `[col, row]` and actions `0=up, 1=down, 2=left, 3=right`.\n- Dynamic `DynamicEnv`-style workflows also use `[col, row]`, but actions are `0=up, 1=down, 2=right, 3=left`; the distilled source agents expose a fifth network output\u002Faction that behaves as a no-op in the dynamic environment.\n- GUI examples are interactive and render-oriented. For validation or CI, use headless logic equivalent to `render_mode=None` rather than opening Pygame loops.\n","# GridWorld DP and Control\n\nUse this sub-skill when the task is about the repository's 5x5 GridWorld workflows: policy iteration, value iteration, tabular SARSA, tabular Q-learning, Deep SARSA, or REINFORCE. It is self-contained; do not depend on the original checkout at runtime.\n\n## Route here for\n\n- Explaining, adapting, or debugging GridWorld dynamic programming (`1-policy_iteration.py`, `2-value_iteration.py`) as provenance workflow labels.\n- Explaining, adapting, or debugging tabular control (`3-sarsa.py`, `4-q_learning.py`) as provenance workflow labels.\n- Making tiny CPU checks for Deep SARSA and REINFORCE neural update logic (`5-deep_sarsa.py`, `6-reinforce.py`) as provenance workflow labels.\n- Diagnosing row\u002Fcolumn versus column\u002Frow coordinate mistakes, action-order mismatches, and render\u002Fheadless behavior.\n\n## Route away from\n\n- Gymnasium CartPole DQN\u002FA2C\u002FPPO: use the CartPole sub-skill.\n- Atari Breakout\u002FPong DQN\u002FPPO, preprocessing, ROMs, or W&B: use the standard Atari sub-skill.\n- Montezuma\u002FPitfall\u002FPrivateEye hard-exploration, RND, Go-Explore, or robustification: use the hard Atari sub-skill.\n\n## Start here\n\n1. Read [`references\u002Falgorithm-guide.md`](references\u002Falgorithm-guide.md) for the MDP layout, coordinate conventions, formulas, API-shaped examples, and safe adaptation patterns.\n2. Read [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md) when a grid policy points in the wrong direction, rendering blocks on a server, tabular values do not update, or Torch neural checks fail.\n3. Run the bundled smoke script when you need a self-contained, non-rendering API check:\n\n```bash\npython scripts\u002Fgrid_world_smoke.py --help\npython scripts\u002Fgrid_world_smoke.py\npython scripts\u002Fgrid_world_smoke.py --section dp\npython scripts\u002Fgrid_world_smoke.py --section neural --strict-torch\n```\n\nThe smoke script intentionally reimplements tiny fixtures and update steps. It does not import source repository files, open Pygame, train to convergence, or require a display.\n\n## Critical facts to preserve\n\n- DP workflows use state as `[row, col]`; drawing helpers usually need `(col, row)`, so swap before rendering arrows or markers.\n- Tabular `Env`-style workflows use state as `[col, row]` and actions `0=up, 1=down, 2=left, 3=right`.\n- Dynamic `DynamicEnv`-style workflows also use `[col, row]`, but actions are `0=up, 1=down, 2=right, 3=left`; the distilled source agents expose a fifth network output\u002Faction that behaves as a no-op in the dynamic environment.\n- GUI examples are interactive and render-oriented. For validation or CI, use headless logic equivalent to `render_mode=None` rather than opening Pygame loops.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Freinforcement-learning\u002Fsub-skills\u002Fgrid-world-dp-and-control",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 grid-world-dp-and-control",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill grid-world-dp-and-control",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill grid-world-dp-and-control","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinforcement-learning\u002Fsub-skills\u002Fgrid-world-dp-and-control\u002FSKILL.md",[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Falgorithm-guide.md","REFERENCE","参考文档",11883,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinforcement-learning\u002Fsub-skills\u002Fgrid-world-dp-and-control\u002Freferences\u002Falgorithm-guide.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",8598,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinforcement-learning\u002Fsub-skills\u002Fgrid-world-dp-and-control\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Fgrid_world_smoke.py","SCRIPT","脚本",23845,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Freinforcement-learning\u002Fsub-skills\u002Fgrid-world-dp-and-control\u002Fscripts\u002Fgrid_world_smoke.py",[63,71,78,85,92,99],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"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":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"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":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"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":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"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":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"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":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"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",[],[]]