[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M796J73657E7HN84EM1XPHXJ":3,"skill-same:01M796J73657E7HN84EM1XPHXJ":70},{"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},"01M796J73657E7HN84EM1XPHXJ","configuration-ui",null,"Guide JSON config editing, smoke configs, CLI override precedence, stage command construction, metrics JSONL inspection, and the Streamlit control panel\u002Fjob manager.","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\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: configuration-ui\ndescription: \"Guide JSON config editing, smoke configs, CLI override precedence,\n  stage command construction, metrics JSONL inspection, and the Streamlit\n  control panel\u002Fjob manager.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# configuration-ui\n\nUse this sub-skill when a future Researcher needs to edit or inspect this repo's\nJSON stage configs, build safe training\u002FUI command lines, use smoke configs,\ninspect metrics JSONL, or operate the Streamlit control panel.\n\n## Read first\n\n- For config precedence, dataclass fields, smoke\u002Ffull JSON behavior, and CLI\n  override rules, read [references\u002Fconfiguration.md](references\u002Fconfiguration.md).\n- For Streamlit pages, job registry behavior, GPU-busy guarding, logs, and UI\n  metrics, read [references\u002Fcontrol-panel-ui.md](references\u002Fcontrol-panel-ui.md).\n- For common config\u002FUI\u002Fmetrics failures, read\n  [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md).\n\n## Safe bundled helpers\n\n- [scripts\u002Fprint_config_summary.py](scripts\u002Fprint_config_summary.py) summarizes\n  one or more JSON config files and compares explicit keys against a base JSON\n  without importing the repo.\n- [scripts\u002Finspect_metrics_jsonl.py](scripts\u002Finspect_metrics_jsonl.py)\n  summarizes metrics JSONL columns, last row, numeric ranges, and\n  malformed-line counts without pandas or repo imports.\n\nBoth helpers are read-only by default, have `--help`, and include `--demo` tiny\nfixture modes.\n\n## Route elsewhere\n\n- Algorithm choices, loss interpretation, RLHF stage sequencing, KL\u002Freward\n  tuning, and training-stage failure analysis: use\n  [..\u002Fpost-training-rlhf\u002FSKILL.md](..\u002Fpost-training-rlhf\u002FSKILL.md).\n- Base architecture, legacy pretraining, DDP\u002Fbf16 pretraining, checkpoint\n  structure, and memory planning: use\n  [..\u002Fmodel-pretraining\u002FSKILL.md](..\u002Fmodel-pretraining\u002FSKILL.md).\n- Dataset schemas, packed HDF5 validation, preference JSONL validation, and RL\n  prompt validation: use [..\u002Fdata-preparation\u002FSKILL.md](..\u002Fdata-preparation\u002FSKILL.md).\n- GSM8K evaluation details, direct chat\u002Fraw inference, checkpoint loading, and\n  sampling controls: use [..\u002Fevaluation-chat\u002FSKILL.md](..\u002Fevaluation-chat\u002FSKILL.md).\n\n## Operating guardrails\n\n- Prefer `configs\u002Fsmoke\u002F*.json` for command\u002Fconfig smoke checks; those use a\n  sibling `configs\u002Fsmoke\u002Fbase.json` and are CPU-oriented.\n- Use `--print-config` before launching long jobs when changing JSON or CLI\n  overrides.\n- Treat UI launches as process-spawning operations: check the UI job registry\n  status and the GPU-busy guard before starting GPU jobs.\n- Treat Weights & Biases as optional. The repo always writes metrics JSONL under\n  the configured log directory even when W&B is disabled or unavailable.\n","# configuration-ui\n\nUse this sub-skill when a future Researcher needs to edit or inspect this repo's\nJSON stage configs, build safe training\u002FUI command lines, use smoke configs,\ninspect metrics JSONL, or operate the Streamlit control panel.\n\n## Read first\n\n- For config precedence, dataclass fields, smoke\u002Ffull JSON behavior, and CLI\n  override rules, read [references\u002Fconfiguration.md](references\u002Fconfiguration.md).\n- For Streamlit pages, job registry behavior, GPU-busy guarding, logs, and UI\n  metrics, read [references\u002Fcontrol-panel-ui.md](references\u002Fcontrol-panel-ui.md).\n- For common config\u002FUI\u002Fmetrics failures, read\n  [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md).\n\n## Safe bundled helpers\n\n- [scripts\u002Fprint_config_summary.py](scripts\u002Fprint_config_summary.py) summarizes\n  one or more JSON config files and compares explicit keys against a base JSON\n  without importing the repo.\n- [scripts\u002Finspect_metrics_jsonl.py](scripts\u002Finspect_metrics_jsonl.py)\n  summarizes metrics JSONL columns, last row, numeric ranges, and\n  malformed-line counts without pandas or repo imports.\n\nBoth helpers are read-only by default, have `--help`, and include `--demo` tiny\nfixture modes.\n\n## Route elsewhere\n\n- Algorithm choices, loss interpretation, RLHF stage sequencing, KL\u002Freward\n  tuning, and training-stage failure analysis: use\n  [..\u002Fpost-training-rlhf\u002FSKILL.md](..\u002Fpost-training-rlhf\u002FSKILL.md).\n- Base architecture, legacy pretraining, DDP\u002Fbf16 pretraining, checkpoint\n  structure, and memory planning: use\n  [..\u002Fmodel-pretraining\u002FSKILL.md](..\u002Fmodel-pretraining\u002FSKILL.md).\n- Dataset schemas, packed HDF5 validation, preference JSONL validation, and RL\n  prompt validation: use [..\u002Fdata-preparation\u002FSKILL.md](..\u002Fdata-preparation\u002FSKILL.md).\n- GSM8K evaluation details, direct chat\u002Fraw inference, checkpoint loading, and\n  sampling controls: use [..\u002Fevaluation-chat\u002FSKILL.md](..\u002Fevaluation-chat\u002FSKILL.md).\n\n## Operating guardrails\n\n- Prefer `configs\u002Fsmoke\u002F*.json` for command\u002Fconfig smoke checks; those use a\n  sibling `configs\u002Fsmoke\u002Fbase.json` and are CPU-oriented.\n- Use `--print-config` before launching long jobs when changing JSON or CLI\n  overrides.\n- Treat UI launches as process-spawning operations: check the UI job registry\n  status and the GPU-busy guard before starting GPU jobs.\n- Treat Weights & Biases as optional. The repo always writes metrics JSONL under\n  the configured log directory even when W&B is disabled or unavailable.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui",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 configuration-ui",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill configuration-ui",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill configuration-ui","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002FSKILL.md",[46,52,56,60,66],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fconfiguration.md","REFERENCE","参考文档",7610,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002Freferences\u002Fconfiguration.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fcontrol-panel-ui.md",6811,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002Freferences\u002Fcontrol-panel-ui.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",7929,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Finspect_metrics_jsonl.py","SCRIPT","脚本",5998,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002Fscripts\u002Finspect_metrics_jsonl.py",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fprint_config_summary.py",6451,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Ftrain-llm-from-scratch\u002Fsub-skills\u002Fconfiguration-ui\u002Fscripts\u002Fprint_config_summary.py",[71,79,86,93,100,107],{"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":75,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"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":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"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":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"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":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"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":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"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":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"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",[],[]]