[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78ZCDPWQ5H72DN2YCJ7HD30":3,"skill-same:01M78ZCDPWQ5H72DN2YCJ7HD30":61},{"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},"01M78ZCDPWQ5H72DN2YCJ7HD30","llama-factory",null,"Use for LlamaFactory repository workflows: training configs, dataset\u002Ftemplate preparation, model loading\u002Fexport, inference\u002FAPI\u002FWeb UI operations, and experimental USE_V1 flows.","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\u002Fllama-factory\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: llama-factory\ndescription: \"Use for LlamaFactory repository workflows: training configs,\n  dataset\u002Ftemplate preparation, model loading\u002Fexport, inference\u002FAPI\u002FWeb UI\n  operations, and experimental USE_V1 flows.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# LlamaFactory\n\nUse this skill when a user asks for help with LlamaFactory (`llamafactory` package, `llamafactory-cli`, or `lmf`) rather than generic Hugging Face training. The default architecture is v0; route to `v1-experimental` only when the user explicitly sets or asks about `USE_V1=1` or v1-style configs.\n\n## Start Here\n\n1. Identify the requested surface: data\u002Ftemplate, training config, model\u002Fexport, inference\u002FAPI, Web UI\u002Fops, or v1.\n2. If the task involves a runnable command, prefer static validation first: inspect config keys, render the intended command, or run a bundled helper with `--help` before starting training, downloads, API servers, or distributed launches.\n3. Confirm optional dependencies and hardware before recommending vLLM, SGLang, DeepSpeed, FlashAttention, bitsandbytes, GPTQ, AWQ, HQQ, EETQ, Unsloth, Liger Kernel, KTransformers, NPU, Ray, FSDP, or Megatron-core paths.\n4. Keep credentials and local machine paths out of reusable examples: hub tokens, tracker keys, private model paths, and dataset paths should be placeholders.\n5. Read `references\u002Ftroubleshooting.md` for cross-cutting install\u002Fimport, dependency, CLI, data\u002Fconfig, backend, and environment-variable issues.\n\n## Route By Task\n\n| User request | Use |\n| --- | --- |\n| Registering datasets, fixing `dataset_info.json`, Alpaca\u002FShareGPT\u002FOpenAI rows, multimodal columns, tool-calling rows, templates, preprocessing, packed data | `sub-skills\u002Fdata-and-templates\u002F` |\n| Writing or debugging `llamafactory-cli train` YAML\u002FJSON, CLI overrides, LoRA\u002FQLoRA\u002Ffull\u002Ffreeze, SFT\u002FPT\u002FRM\u002FPPO\u002FDPO\u002FKTO, distributed launch, logging | `sub-skills\u002Ftraining-and-configs\u002F` |\n| Using `ChatModel`, `chat`, `webchat`, `api`, OpenAI-compatible endpoints, streaming, scores, `infer_backend`, vLLM or SGLang inference | `sub-skills\u002Finference-and-serving\u002F` |\n| Loading tokenizers\u002Fmodels\u002Fprocessors, adapter setup, LoRA merge\u002Fexport, quantization, checkpoint conversion\u002Finit utilities, hub mirror behavior | `sub-skills\u002Fmodel-loading-and-export\u002F` |\n| Installing\u002Frunning LlamaFactory, `help`, `version`, `env`, LlamaBoard `webui`, operational env vars, monitors, Docker\u002Fbackend notes | `sub-skills\u002Fwebui-and-ops\u002F` |\n| Experimental v1 trainer\u002Fcore\u002Fplugin\u002Fconfig flows selected with `USE_V1=1` | `sub-skills\u002Fv1-experimental\u002F` |\n\n## CLI And Package Facts\n\n- Package\u002Fdistribution: `llamafactory`.\n- Public version in this repo snapshot: `0.9.6.dev0`.\n- Console scripts: `llamafactory-cli` and `lmf`, both dispatch to `llamafactory.cli:main`.\n- v0 CLI routes include `train`, `chat`, `api`, `export`, `webchat`, `webui`, `env`, `version`, and `help`.\n- Multi-GPU training may be wrapped in `torchrun` automatically when multiple devices are visible, or forced with `FORCE_TORCHRUN=1`.\n- `USE_V1=1` switches the launcher to the experimental v1 architecture and changes supported commands\u002Fconfig shape.\n\n## Minimal Checks\n\nUse these checks after installing LlamaFactory in a real environment:\n\n```bash\npython - \u003C\u003C'PY'\nimport llamafactory\nprint(llamafactory.__version__)\nPY\nllamafactory-cli help\n```\n\nIf the package is only partially installed, root imports may succeed while workflow imports fail later. Run the sanity helper in `sub-skills\u002Fwebui-and-ops\u002Fscripts\u002Fllamafactory_sanity_check.py` to distinguish metadata\u002Fimport\u002FCLI\u002Fdependency problems.\n\n## Bundled Helpers\n\n- `sub-skills\u002Fdata-and-templates\u002Fscripts\u002Fvalidate_dataset_entry.py` checks dataset registry entries and tiny row files without importing LlamaFactory.\n- `sub-skills\u002Ftraining-and-configs\u002Fscripts\u002Frender_train_command.py` merges simple YAML\u002FJSON overrides and prints a safe training command.\n- `sub-skills\u002Ftraining-and-configs\u002Fscripts\u002Feval_bleu_rouge.py` provides dependency-light prediction metric checks for generated JSONL outputs.\n- `sub-skills\u002Finference-and-serving\u002Fscripts\u002Fopenai_api_smoke.py` probes an OpenAI-compatible local API when the user supplies server details.\n- `sub-skills\u002Finference-and-serving\u002Fscripts\u002Fvllm_batch_infer.py` wraps batch inference intent and requires the caller to provide an installed inference environment.\n- `sub-skills\u002Fmodel-loading-and-export\u002Fscripts\u002Fcheck_export_config.py` flags static export\u002Fmerge\u002Fquantization config pitfalls.\n- `sub-skills\u002Fwebui-and-ops\u002Fscripts\u002Fllamafactory_sanity_check.py` checks Python\u002Fpackage\u002FCLI availability and common optional modules.\n- `sub-skills\u002Fv1-experimental\u002Fscripts\u002Fcheck_v1_config_keys.py` warns about mixed v0\u002Fv1 config keys without importing LlamaFactory.\n\n## Repo-Level References\n\n- `references\u002Frepo-provenance.md` records the source snapshot and evidence paths used to build this skill.\n- `references\u002Ftroubleshooting.md` covers cross-cutting failures and points to the nearest owning sub-skill.\n\n## Safety Notes\n\n- Do not run model downloads, training, distributed launches, API servers, Web UI sessions, checkpoint conversion, or export jobs unless the user asks and the environment\u002Fhardware are suitable.\n- Treat repo examples and tests as evidence, not runtime dependencies for this skill. The bundled references and scripts are the self-contained runtime material.\n- For dependency-heavy native verification, use a full LlamaFactory installation with PyTorch and the selected optional backend packages.\n","# LlamaFactory\n\nUse this skill when a user asks for help with LlamaFactory (`llamafactory` package, `llamafactory-cli`, or `lmf`) rather than generic Hugging Face training. The default architecture is v0; route to `v1-experimental` only when the user explicitly sets or asks about `USE_V1=1` or v1-style configs.\n\n## Start Here\n\n1. Identify the requested surface: data\u002Ftemplate, training config, model\u002Fexport, inference\u002FAPI, Web UI\u002Fops, or v1.\n2. If the task involves a runnable command, prefer static validation first: inspect config keys, render the intended command, or run a bundled helper with `--help` before starting training, downloads, API servers, or distributed launches.\n3. Confirm optional dependencies and hardware before recommending vLLM, SGLang, DeepSpeed, FlashAttention, bitsandbytes, GPTQ, AWQ, HQQ, EETQ, Unsloth, Liger Kernel, KTransformers, NPU, Ray, FSDP, or Megatron-core paths.\n4. Keep credentials and local machine paths out of reusable examples: hub tokens, tracker keys, private model paths, and dataset paths should be placeholders.\n5. Read `references\u002Ftroubleshooting.md` for cross-cutting install\u002Fimport, dependency, CLI, data\u002Fconfig, backend, and environment-variable issues.\n\n## Route By Task\n\n| User request | Use |\n| --- | --- |\n| Registering datasets, fixing `dataset_info.json`, Alpaca\u002FShareGPT\u002FOpenAI rows, multimodal columns, tool-calling rows, templates, preprocessing, packed data | `sub-skills\u002Fdata-and-templates\u002F` |\n| Writing or debugging `llamafactory-cli train` YAML\u002FJSON, CLI overrides, LoRA\u002FQLoRA\u002Ffull\u002Ffreeze, SFT\u002FPT\u002FRM\u002FPPO\u002FDPO\u002FKTO, distributed launch, logging | `sub-skills\u002Ftraining-and-configs\u002F` |\n| Using `ChatModel`, `chat`, `webchat`, `api`, OpenAI-compatible endpoints, streaming, scores, `infer_backend`, vLLM or SGLang inference | `sub-skills\u002Finference-and-serving\u002F` |\n| Loading tokenizers\u002Fmodels\u002Fprocessors, adapter setup, LoRA merge\u002Fexport, quantization, checkpoint conversion\u002Finit utilities, hub mirror behavior | `sub-skills\u002Fmodel-loading-and-export\u002F` |\n| Installing\u002Frunning LlamaFactory, `help`, `version`, `env`, LlamaBoard `webui`, operational env vars, monitors, Docker\u002Fbackend notes | `sub-skills\u002Fwebui-and-ops\u002F` |\n| Experimental v1 trainer\u002Fcore\u002Fplugin\u002Fconfig flows selected with `USE_V1=1` | `sub-skills\u002Fv1-experimental\u002F` |\n\n## CLI And Package Facts\n\n- Package\u002Fdistribution: `llamafactory`.\n- Public version in this repo snapshot: `0.9.6.dev0`.\n- Console scripts: `llamafactory-cli` and `lmf`, both dispatch to `llamafactory.cli:main`.\n- v0 CLI routes include `train`, `chat`, `api`, `export`, `webchat`, `webui`, `env`, `version`, and `help`.\n- Multi-GPU training may be wrapped in `torchrun` automatically when multiple devices are visible, or forced with `FORCE_TORCHRUN=1`.\n- `USE_V1=1` switches the launcher to the experimental v1 architecture and changes supported commands\u002Fconfig shape.\n\n## Minimal Checks\n\nUse these checks after installing LlamaFactory in a real environment:\n\n```bash\npython - \u003C\u003C'PY'\nimport llamafactory\nprint(llamafactory.__version__)\nPY\nllamafactory-cli help\n```\n\nIf the package is only partially installed, root imports may succeed while workflow imports fail later. Run the sanity helper in `sub-skills\u002Fwebui-and-ops\u002Fscripts\u002Fllamafactory_sanity_check.py` to distinguish metadata\u002Fimport\u002FCLI\u002Fdependency problems.\n\n## Bundled Helpers\n\n- `sub-skills\u002Fdata-and-templates\u002Fscripts\u002Fvalidate_dataset_entry.py` checks dataset registry entries and tiny row files without importing LlamaFactory.\n- `sub-skills\u002Ftraining-and-configs\u002Fscripts\u002Frender_train_command.py` merges simple YAML\u002FJSON overrides and prints a safe training command.\n- `sub-skills\u002Ftraining-and-configs\u002Fscripts\u002Feval_bleu_rouge.py` provides dependency-light prediction metric checks for generated JSONL outputs.\n- `sub-skills\u002Finference-and-serving\u002Fscripts\u002Fopenai_api_smoke.py` probes an OpenAI-compatible local API when the user supplies server details.\n- `sub-skills\u002Finference-and-serving\u002Fscripts\u002Fvllm_batch_infer.py` wraps batch inference intent and requires the caller to provide an installed inference environment.\n- `sub-skills\u002Fmodel-loading-and-export\u002Fscripts\u002Fcheck_export_config.py` flags static export\u002Fmerge\u002Fquantization config pitfalls.\n- `sub-skills\u002Fwebui-and-ops\u002Fscripts\u002Fllamafactory_sanity_check.py` checks Python\u002Fpackage\u002FCLI availability and common optional modules.\n- `sub-skills\u002Fv1-experimental\u002Fscripts\u002Fcheck_v1_config_keys.py` warns about mixed v0\u002Fv1 config keys without importing LlamaFactory.\n\n## Repo-Level References\n\n- `references\u002Frepo-provenance.md` records the source snapshot and evidence paths used to build this skill.\n- `references\u002Ftroubleshooting.md` covers cross-cutting failures and points to the nearest owning sub-skill.\n\n## Safety Notes\n\n- Do not run model downloads, training, distributed launches, API servers, Web UI sessions, checkpoint conversion, or export jobs unless the user asks and the environment\u002Fhardware are suitable.\n- Treat repo examples and tests as evidence, not runtime dependencies for this skill. The bundled references and scripts are the self-contained runtime material.\n- For dependency-heavy native verification, use a full LlamaFactory installation with PyTorch and the selected optional backend packages.\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\u002Fllama-factory",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 llama-factory",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill llama-factory",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill llama-factory","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fllama-factory\u002FSKILL.md",[47,53,57],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Frepo-provenance.md","REFERENCE","参考文档",1564,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fllama-factory\u002Freferences\u002Frepo-provenance.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Frepo-routing-metadata.json",567,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fllama-factory\u002Freferences\u002Frepo-routing-metadata.json",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Ftroubleshooting.md",4182,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fllama-factory\u002Freferences\u002Ftroubleshooting.md",[62,69,76,83,90,97],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"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":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"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":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},"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":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},"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":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},"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":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},"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",[],[]]