[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78V1EJS135JXYBJW6B4P7RT":3,"skill-same:01M78V1EJS135JXYBJW6B4P7RT":74},{"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},"01M78V1EJS135JXYBJW6B4P7RT","modeling-and-tokenization",null,"Use when inspecting ColBERT checkpoints, ColBERTConfig model\u002Ftokenizer fields, QueryTokenizer or DocTokenizer behavior, query\u002Fdocument max lengths, marker tokens, punctuation masking, embedding dimensions, or safe model\u002Ftokenizer import diagnostics.","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\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: modeling-and-tokenization\ndescription: \"Use when inspecting ColBERT checkpoints, ColBERTConfig\n  model\u002Ftokenizer fields, QueryTokenizer or DocTokenizer behavior,\n  query\u002Fdocument max lengths, marker tokens, punctuation masking, embedding\n  dimensions, or safe model\u002Ftokenizer import diagnostics.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Modeling and Tokenization\n\nUse this sub-skill when an agent needs to reason about ColBERT model checkpoints, tokenizer behavior, config compatibility, text-to-token tensorization, or query\u002Fdocument embedding shapes before indexing, searching, or training.\n\n## Route Here For\n\n- Inspecting `Checkpoint(name, colbert_config=None, verbose=3)` loading behavior and `checkpoint.colbert_config` values.\n- Changing or debugging `ColBERTConfig` fields such as `query_maxlen`, `doc_maxlen`, `dim`, `mask_punctuation`, `attend_to_mask_tokens`, `query_token_id`, or `doc_token_id`.\n- Understanding how `QueryTokenizer` and `DocTokenizer` insert `[Q]` and `[D]` markers, truncate text, pad queries with `[MASK]`, and return IDs\u002Fmasks.\n- Explaining `Checkpoint.queryFromText()` and `Checkpoint.docFromText()` output shapes, `to_cpu`, `keep_dims`, `return_tokens`, and packed document outputs.\n- Diagnosing checkpoint metadata, tokenizer downloads, local-vs-Hugging Face model names, CPU\u002FGPU model loading caveats, or missing optional dependencies.\n\n## Start With\n\n1. Read `references\u002Fmodeling-and-tokenization.md` for the conceptual model, checkpoint\u002Fconfig merge behavior, marker tokens, tokenization, and embedding outputs.\n2. Read `references\u002Fconfig-reference.md` before changing shape-sensitive fields such as `query_maxlen`, `doc_maxlen`, or `dim`.\n3. Read `references\u002Fapi-reference.md` for verified imports, constructor signatures, tensorizer methods, encoding outputs, and shape invariants.\n4. Use `scripts\u002Finspect_checkpoint_config.py` when you need a safe ColBERTConfig summary; pass `--allow-remote` only when remote Hugging Face access is intentional.\n5. Use `scripts\u002Ftokenization_smoke.py` when a local checkpoint or explicitly allowed model name should be checked for marker placement, truncation, and tensor shapes.\n6. Read `references\u002Ftroubleshooting.md` when imports, checkpoint metadata, Hugging Face downloads, max-length truncation, punctuation masking, or device behavior is surprising.\n\n## Boundaries\n\n- This sub-skill covers modeling, config, checkpoint, tokenizer, and embedding-inspection tasks.\n- For full indexing\u002Fsearch orchestration, index roots, ranking, or `Indexer`\u002F`Searcher` usage, route to the indexing-and-search sub-skill.\n- For training, triples, distillation, optimizer settings, checkpoint creation, or `Trainer`, route to the training-and-distillation sub-skill.\n- For collection\u002Fquery\u002Franking file schemas beyond text passed into tokenizers, route to the data-and-evaluation sub-skill.\n\n## Operating Cautions\n\n- Prefer local checkpoint directories for deterministic inspection. Hugging Face model names such as `colbert-ir\u002Fcolbertv2.0` or `bert-base-uncased` can require network\u002Fcache access.\n- `ColBERTConfig.load_from_checkpoint()` can return `None` for a generic Hugging Face backbone or a checkpoint directory missing ColBERT `artifact.metadata`.\n- Query and document marker IDs are inserted at column `1`; do not pre-insert marker IDs before calling ColBERT tensorizers.\n- Query tensor width is normally `query_maxlen`; document tensor width is capped by `doc_maxlen` but may be the longest sequence in the batch, not always exactly `doc_maxlen`.\n- CPU-only import\u002Fconfig inspection is safe; full model loading can compile a small CPU extension, download tokenizer\u002Fmodel files, or hit torch\u002Ftransformers compatibility issues.\n","# Modeling and Tokenization\n\nUse this sub-skill when an agent needs to reason about ColBERT model checkpoints, tokenizer behavior, config compatibility, text-to-token tensorization, or query\u002Fdocument embedding shapes before indexing, searching, or training.\n\n## Route Here For\n\n- Inspecting `Checkpoint(name, colbert_config=None, verbose=3)` loading behavior and `checkpoint.colbert_config` values.\n- Changing or debugging `ColBERTConfig` fields such as `query_maxlen`, `doc_maxlen`, `dim`, `mask_punctuation`, `attend_to_mask_tokens`, `query_token_id`, or `doc_token_id`.\n- Understanding how `QueryTokenizer` and `DocTokenizer` insert `[Q]` and `[D]` markers, truncate text, pad queries with `[MASK]`, and return IDs\u002Fmasks.\n- Explaining `Checkpoint.queryFromText()` and `Checkpoint.docFromText()` output shapes, `to_cpu`, `keep_dims`, `return_tokens`, and packed document outputs.\n- Diagnosing checkpoint metadata, tokenizer downloads, local-vs-Hugging Face model names, CPU\u002FGPU model loading caveats, or missing optional dependencies.\n\n## Start With\n\n1. Read `references\u002Fmodeling-and-tokenization.md` for the conceptual model, checkpoint\u002Fconfig merge behavior, marker tokens, tokenization, and embedding outputs.\n2. Read `references\u002Fconfig-reference.md` before changing shape-sensitive fields such as `query_maxlen`, `doc_maxlen`, or `dim`.\n3. Read `references\u002Fapi-reference.md` for verified imports, constructor signatures, tensorizer methods, encoding outputs, and shape invariants.\n4. Use `scripts\u002Finspect_checkpoint_config.py` when you need a safe ColBERTConfig summary; pass `--allow-remote` only when remote Hugging Face access is intentional.\n5. Use `scripts\u002Ftokenization_smoke.py` when a local checkpoint or explicitly allowed model name should be checked for marker placement, truncation, and tensor shapes.\n6. Read `references\u002Ftroubleshooting.md` when imports, checkpoint metadata, Hugging Face downloads, max-length truncation, punctuation masking, or device behavior is surprising.\n\n## Boundaries\n\n- This sub-skill covers modeling, config, checkpoint, tokenizer, and embedding-inspection tasks.\n- For full indexing\u002Fsearch orchestration, index roots, ranking, or `Indexer`\u002F`Searcher` usage, route to the indexing-and-search sub-skill.\n- For training, triples, distillation, optimizer settings, checkpoint creation, or `Trainer`, route to the training-and-distillation sub-skill.\n- For collection\u002Fquery\u002Franking file schemas beyond text passed into tokenizers, route to the data-and-evaluation sub-skill.\n\n## Operating Cautions\n\n- Prefer local checkpoint directories for deterministic inspection. Hugging Face model names such as `colbert-ir\u002Fcolbertv2.0` or `bert-base-uncased` can require network\u002Fcache access.\n- `ColBERTConfig.load_from_checkpoint()` can return `None` for a generic Hugging Face backbone or a checkpoint directory missing ColBERT `artifact.metadata`.\n- Query and document marker IDs are inserted at column `1`; do not pre-insert marker IDs before calling ColBERT tensorizers.\n- Query tensor width is normally `query_maxlen`; document tensor width is capped by `doc_maxlen` but may be the longest sequence in the batch, not always exactly `doc_maxlen`.\n- CPU-only import\u002Fconfig inspection is safe; full model loading can compile a small CPU extension, download tokenizer\u002Fmodel files, or hit torch\u002Ftransformers compatibility issues.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization",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 modeling-and-tokenization",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill modeling-and-tokenization",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill modeling-and-tokenization","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002FSKILL.md",[46,52,56,60,64,70],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",6871,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fconfig-reference.md",6071,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002Freferences\u002Fconfig-reference.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fmodeling-and-tokenization.md",6667,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002Freferences\u002Fmodeling-and-tokenization.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Ftroubleshooting.md",7156,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002Freferences\u002Ftroubleshooting.md",{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Finspect_checkpoint_config.py","SCRIPT","脚本",3881,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002Fscripts\u002Finspect_checkpoint_config.py",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Ftokenization_smoke.py",5801,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fcolbert\u002Fsub-skills\u002Fmodeling-and-tokenization\u002Fscripts\u002Ftokenization_smoke.py",[75,83,90,97,104,111],{"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":79,"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.","Apache-2.0","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":79,"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":79,"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":79,"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":79,"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":79,"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",[],[]]