[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78VGXHEP800QPFPK59PZT36":3,"skill-same:01M78VGXHEP800QPFPK59PZT36":75},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M78VGXHEP800QPFPK59PZT36","cli-and-studio",null,"Guides DataChain CLI commands, dataset and storage command routing, Studio auth\u002Fjobs\u002Fpipelines, and environment configuration.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: cli-and-studio\ndescription: \"Guides DataChain CLI commands, dataset and storage command\n  routing, Studio auth\u002Fjobs\u002Fpipelines, and environment configuration.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# CLI and Studio\n\nUse this sub-skill when the user asks how to choose or run `datachain` CLI\ncommands, diagnose command parser behavior, route storage versus dataset\noperations, authenticate with DataChain Studio, run or monitor Studio jobs,\nmanage Studio pipelines, configure CLI\u002FStudio environment variables, or inspect\nDataChain bundled skill install targets.\n\n## Trigger Phrases\n\nLoad this sub-skill for prompts containing or implying:\n\n- `datachain --help`, `datachain -V`, verbose\u002Fquiet behavior, parser errors, or\n  command-specific `--help` output;\n- storage commands: `bucket status`, `cp`, `clone`, `ls`, `du`, `find`, `index`,\n  public\u002Fprivate buckets, `--anon`, `--studio`, `--local`, or `--all`;\n- dataset\u002Fdisplay commands: `dataset`, `ds`, `dataset ls`, `dataset rm`,\n  `dataset remove`, `dataset edit`, `dataset pull`, or `show`;\n- maintenance commands: `gc`, `clear-cache`, outdated checkpoints, temporary\n  tables, or local cache cleanup;\n- Studio commands: `auth login`, `auth token`, `auth team`, `auth logout`,\n  `job run`, `job logs`, `job cancel`, `job clusters`, `pipeline create`,\n  `pipeline list`, `pipeline pause`, `pipeline resume`, `pipeline status`, or\n  `pipeline remove-job`;\n- `datachain skill install`, `datachain skill list`, `--target`, `--local`,\n  Claude\u002FCursor\u002FCodex\u002FPi\u002FCopilot target selection, and safe layout previews;\n- `DATACHAIN_ROOT_DIR`, `DATACHAIN_GLOBAL_CONFIG_DIR`,\n  `DATACHAIN_SYSTEM_CONFIG_DIR`, `DATACHAIN_NO_ANALYTICS`,\n  `DATACHAIN_STUDIO_URL`, `DATACHAIN_STUDIO_TOKEN`, `DATACHAIN_STUDIO_TEAM`,\n  `DATACHAIN_NAMESPACE`, `DATACHAIN_PROJECT`, or\n  `DATACHAIN_IGNORE_CHECKPOINTS`.\n\n## First Decision\n\n1. **Need a command catalog or parser default** → read\n   [CLI reference](references\u002Fcli-reference.md). If the user only needs to\n   verify help output safely, run\n   [cli_help_smoke.py](scripts\u002Fcli_help_smoke.py) with a command path such as\n   `job run` or `dataset ls`; the helper appends `--help` and avoids credentialed\n   operations.\n2. **Need Studio auth, jobs, scheduling, logs, clusters, or pipelines** → read\n   [Studio and jobs](references\u002Fstudio-and-jobs.md). Confirm token\u002Fteam and\n   scheduling intent before running any Studio mutation.\n3. **Need environment variables, config precedence, namespace\u002Fproject defaults,\n   or token handling** → read [configuration](references\u002Fconfiguration.md).\n4. **Need error diagnosis** → read\n   [troubleshooting](references\u002Ftroubleshooting.md), especially for missing\n   auth, local-vs-Studio flavor confusion, storage access, `show`\u002Fschema\n   confusion, job-log cancellation misconceptions, scheduled jobs, and parser\n   errors.\n5. **Need DataChain bundled skill target layouts** → use\n   [skill_layout_check.py](scripts\u002Fskill_layout_check.py) for a dry-run layout\n   preview, then route detailed target-layout and installed-skill behavior to\n   sibling [`agent-harness`](..\u002Fagent-harness\u002FSKILL.md).\n\n## Command Routing Rules\n\n- Use **storage commands** when the subject is a URI\u002Fpath or object-storage\n  listing\u002Fcopy\u002Findex task (`s3:\u002F\u002F`, `gs:\u002F\u002F`, `az:\u002F\u002F`, `hf:\u002F\u002F`, `file:\u002F\u002F`, local\n  paths). Storage `ls` is not the same as `dataset ls`.\n- Use **dataset commands** when the subject is a named saved DataChain dataset\n  and version (`name`, `namespace.project.name`, or `name@version`). Use `show`\n  to inspect dataset rows, schema, or query script.\n- Use **Studio flavor flags** deliberately. For `ls` and `dataset ls`, no flavor\n  flag defaults to local. `--studio` requires Studio auth; `--all` lists both\n  only when a token is available and otherwise falls back to local.\n- Use **Studio job commands** for remote execution of a query file. `job run`\n  submits work to Studio and may keep running after local log output stops.\n- Use **pipeline commands** for Studio-managed dependency pipelines over dataset\n  updates. Pipelines are separate from Python SDK pipeline authoring.\n- Use **skill commands** only for DataChain's bundled agent-skill installer; the\n  knowledge-base methodology and installed agent skill content are owned by\n  `agent-harness`.\n\n## Safety Rules\n\n- Prefer `datachain \u003Ccommand> --help` or the bundled help smoke helper before\n  composing unfamiliar commands.\n- Treat these as mutating: `cp`, `clone`, `index`, `dataset pull`, `dataset edit`,\n  `dataset rm\u002Fremove`, `gc`, `clear-cache`, `auth login\u002Flogout\u002Fteam`, `job run`,\n  `job cancel`, `pipeline create\u002Fpause\u002Fresume\u002Fremove-job`, and\n  `skill install\u002Funinstall`.\n- Before a mutating command, confirm the dataset name\u002Fversion, output directory,\n  Studio team, target agent, and whether the operation is local or Studio-scoped.\n- Do not print, paste, commit, or store real Studio tokens, cloud credentials, or\n  secret `--env` values. `datachain auth token` intentionally prints a secret;\n  use it only when the user explicitly asks to inspect or copy the token.\n- For public buckets, `--anon` is acceptable. For private storage, verify the\n  credential source and consider `bucket status` before listing or copying.\n- Closing `job run` or `job logs` output does not cancel a Studio job. Use\n  `datachain job cancel \u003Cid>` for cancellation.\n- Do not route Python SDK chain construction, UDF signatures, `.save()`, or\n  `read_storage` code patterns here; send those to `..\u002Fsdk-pipelines\u002F`.\n- Do not route query expression operators\u002Ffunctions here; send those to\n  `..\u002Fquery-engine\u002F`.\n- Do not route repo contributor testing, backend matrix development, or release\n  work here; send those to `..\u002Frepo-development\u002F`.\n\n## Native Verification Candidates Owned Here\n\n- Parser and flavor defaults: CLI parsing, `-v\u002F-q`, `bucket status`, and\n  `determine_flavors` unit coverage.\n- Skill command behavior: bundled skill list\u002Finstall target parsing and layout\n  rules; use the layout helper instead of performing real installs during\n  usability checks.\n- CLI end-to-end candidates: storage `ls\u002Fdu\u002Ffind\u002Fcp\u002Fclone`, dataset `ls\u002Fedit\u002Frm`,\n  and `gc`; mark network\u002Fobject-copying steps as side-effecting unless run in a\n  controlled fixture.\n- Studio candidates: auth login\u002Flogout\u002Ftoken\u002Fteam, dataset Studio edit\u002Fremove,\n  job run\u002Flog\u002Fcancel\u002Fclusters, reconnect behavior, checkpoint reuse, and\n  scheduled-job behavior with mocked Studio responses.\n","# CLI and Studio\n\nUse this sub-skill when the user asks how to choose or run `datachain` CLI\ncommands, diagnose command parser behavior, route storage versus dataset\noperations, authenticate with DataChain Studio, run or monitor Studio jobs,\nmanage Studio pipelines, configure CLI\u002FStudio environment variables, or inspect\nDataChain bundled skill install targets.\n\n## Trigger Phrases\n\nLoad this sub-skill for prompts containing or implying:\n\n- `datachain --help`, `datachain -V`, verbose\u002Fquiet behavior, parser errors, or\n  command-specific `--help` output;\n- storage commands: `bucket status`, `cp`, `clone`, `ls`, `du`, `find`, `index`,\n  public\u002Fprivate buckets, `--anon`, `--studio`, `--local`, or `--all`;\n- dataset\u002Fdisplay commands: `dataset`, `ds`, `dataset ls`, `dataset rm`,\n  `dataset remove`, `dataset edit`, `dataset pull`, or `show`;\n- maintenance commands: `gc`, `clear-cache`, outdated checkpoints, temporary\n  tables, or local cache cleanup;\n- Studio commands: `auth login`, `auth token`, `auth team`, `auth logout`,\n  `job run`, `job logs`, `job cancel`, `job clusters`, `pipeline create`,\n  `pipeline list`, `pipeline pause`, `pipeline resume`, `pipeline status`, or\n  `pipeline remove-job`;\n- `datachain skill install`, `datachain skill list`, `--target`, `--local`,\n  Claude\u002FCursor\u002FCodex\u002FPi\u002FCopilot target selection, and safe layout previews;\n- `DATACHAIN_ROOT_DIR`, `DATACHAIN_GLOBAL_CONFIG_DIR`,\n  `DATACHAIN_SYSTEM_CONFIG_DIR`, `DATACHAIN_NO_ANALYTICS`,\n  `DATACHAIN_STUDIO_URL`, `DATACHAIN_STUDIO_TOKEN`, `DATACHAIN_STUDIO_TEAM`,\n  `DATACHAIN_NAMESPACE`, `DATACHAIN_PROJECT`, or\n  `DATACHAIN_IGNORE_CHECKPOINTS`.\n\n## First Decision\n\n1. **Need a command catalog or parser default** → read\n   [CLI reference](references\u002Fcli-reference.md). If the user only needs to\n   verify help output safely, run\n   [cli_help_smoke.py](scripts\u002Fcli_help_smoke.py) with a command path such as\n   `job run` or `dataset ls`; the helper appends `--help` and avoids credentialed\n   operations.\n2. **Need Studio auth, jobs, scheduling, logs, clusters, or pipelines** → read\n   [Studio and jobs](references\u002Fstudio-and-jobs.md). Confirm token\u002Fteam and\n   scheduling intent before running any Studio mutation.\n3. **Need environment variables, config precedence, namespace\u002Fproject defaults,\n   or token handling** → read [configuration](references\u002Fconfiguration.md).\n4. **Need error diagnosis** → read\n   [troubleshooting](references\u002Ftroubleshooting.md), especially for missing\n   auth, local-vs-Studio flavor confusion, storage access, `show`\u002Fschema\n   confusion, job-log cancellation misconceptions, scheduled jobs, and parser\n   errors.\n5. **Need DataChain bundled skill target layouts** → use\n   [skill_layout_check.py](scripts\u002Fskill_layout_check.py) for a dry-run layout\n   preview, then route detailed target-layout and installed-skill behavior to\n   sibling [`agent-harness`](..\u002Fagent-harness\u002FSKILL.md).\n\n## Command Routing Rules\n\n- Use **storage commands** when the subject is a URI\u002Fpath or object-storage\n  listing\u002Fcopy\u002Findex task (`s3:\u002F\u002F`, `gs:\u002F\u002F`, `az:\u002F\u002F`, `hf:\u002F\u002F`, `file:\u002F\u002F`, local\n  paths). Storage `ls` is not the same as `dataset ls`.\n- Use **dataset commands** when the subject is a named saved DataChain dataset\n  and version (`name`, `namespace.project.name`, or `name@version`). Use `show`\n  to inspect dataset rows, schema, or query script.\n- Use **Studio flavor flags** deliberately. For `ls` and `dataset ls`, no flavor\n  flag defaults to local. `--studio` requires Studio auth; `--all` lists both\n  only when a token is available and otherwise falls back to local.\n- Use **Studio job commands** for remote execution of a query file. `job run`\n  submits work to Studio and may keep running after local log output stops.\n- Use **pipeline commands** for Studio-managed dependency pipelines over dataset\n  updates. Pipelines are separate from Python SDK pipeline authoring.\n- Use **skill commands** only for DataChain's bundled agent-skill installer; the\n  knowledge-base methodology and installed agent skill content are owned by\n  `agent-harness`.\n\n## Safety Rules\n\n- Prefer `datachain \u003Ccommand> --help` or the bundled help smoke helper before\n  composing unfamiliar commands.\n- Treat these as mutating: `cp`, `clone`, `index`, `dataset pull`, `dataset edit`,\n  `dataset rm\u002Fremove`, `gc`, `clear-cache`, `auth login\u002Flogout\u002Fteam`, `job run`,\n  `job cancel`, `pipeline create\u002Fpause\u002Fresume\u002Fremove-job`, and\n  `skill install\u002Funinstall`.\n- Before a mutating command, confirm the dataset name\u002Fversion, output directory,\n  Studio team, target agent, and whether the operation is local or Studio-scoped.\n- Do not print, paste, commit, or store real Studio tokens, cloud credentials, or\n  secret `--env` values. `datachain auth token` intentionally prints a secret;\n  use it only when the user explicitly asks to inspect or copy the token.\n- For public buckets, `--anon` is acceptable. For private storage, verify the\n  credential source and consider `bucket status` before listing or copying.\n- Closing `job run` or `job logs` output does not cancel a Studio job. Use\n  `datachain job cancel \u003Cid>` for cancellation.\n- Do not route Python SDK chain construction, UDF signatures, `.save()`, or\n  `read_storage` code patterns here; send those to `..\u002Fsdk-pipelines\u002F`.\n- Do not route query expression operators\u002Ffunctions here; send those to\n  `..\u002Fquery-engine\u002F`.\n- Do not route repo contributor testing, backend matrix development, or release\n  work here; send those to `..\u002Frepo-development\u002F`.\n\n## Native Verification Candidates Owned Here\n\n- Parser and flavor defaults: CLI parsing, `-v\u002F-q`, `bucket status`, and\n  `determine_flavors` unit coverage.\n- Skill command behavior: bundled skill list\u002Finstall target parsing and layout\n  rules; use the layout helper instead of performing real installs during\n  usability checks.\n- CLI end-to-end candidates: storage `ls\u002Fdu\u002Ffind\u002Fcp\u002Fclone`, dataset `ls\u002Fedit\u002Frm`,\n  and `gc`; mark network\u002Fobject-copying steps as side-effecting unless run in a\n  controlled fixture.\n- Studio candidates: auth login\u002Flogout\u002Ftoken\u002Fteam, dataset Studio edit\u002Fremove,\n  job run\u002Flog\u002Fcancel\u002Fclusters, reconnect behavior, checkpoint reuse, and\n  scheduled-job behavior with mocked Studio responses.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill cli-and-studio",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill cli-and-studio",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill cli-and-studio","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002FSKILL.md",[47,53,57,61,65,71],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fcli-reference.md","REFERENCE","参考文档",10979,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002Freferences\u002Fcli-reference.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fconfiguration.md",5937,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002Freferences\u002Fconfiguration.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fstudio-and-jobs.md",10651,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002Freferences\u002Fstudio-and-jobs.md",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Ftroubleshooting.md",8821,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002Freferences\u002Ftroubleshooting.md",{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Fcli_help_smoke.py","SCRIPT","脚本",8461,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002Fscripts\u002Fcli_help_smoke.py",{"path":72,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":73,"sourceHtmlUrl":74},"scripts\u002Fskill_layout_check.py",4984,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdatachain\u002Fsub-skills\u002Fcli-and-studio\u002Fscripts\u002Fskill_layout_check.py",[76,83,90,97,104,111],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"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.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"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":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"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":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"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":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"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":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"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",[],[]]