[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78WPX1QQ8553KJSV50TD6RT":3,"skill-same:01M78WPX1QQ8553KJSV50TD6RT":71},{"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},"01M78WPX1QQ8553KJSV50TD6RT","equinox",null,"Operating skill for the Equinox JAX neural-network and PyTree library.","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\u002Fequinox\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: equinox\ndescription: \"Operating skill for the Equinox JAX neural-network and PyTree library.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: Apache 2.0\n---\n\n# Equinox\n\nEquinox is a JAX library for building models as PyTrees, applying filtered JAX\ntransformations, and using reusable neural-network layers, stateful modules,\nand advanced debugging utilities.\n\nUse this skill when the user mentions:\n\n- `equinox`, `eqx`, `eqx.Module`, or `equinox.nn`\n- filtered transforms such as `filter_jit`, `filter_grad`, `filter_vmap`, or\n  `filter_pmap`\n- model surgery, PyTree partitioning, serialization, or runtime errors\n- `Sequential`, `MLP`, `BatchNorm`, `State`, `Shared`, or `inference_mode`\n- `equinox.internal` helpers such as `noinline`, `while_loop`, or `scan`\n\n## Quick start\n\nInstall the public package with:\n\n```bash\npip install equinox\n```\n\nFor a local checkout or a health check, run the bundled [`scripts\u002Fsmoke.py`](scripts\u002Fsmoke.py)\nfrom the generated skill root (the directory containing this `SKILL.md`). Set\n`EQUINOX_SKILL_ROOT` to that absolute directory and keep it as the command\nworking directory:\n\n```bash\ncd \"$EQUINOX_SKILL_ROOT\" && python scripts\u002Fsmoke.py --mode all\n```\n\nIf you need multi-device CPU coverage for `pmap` or sharding checks, add\n`--two-cpu-devices` to the smoke script.\n\n## Route map\n\n### [`sub-skills\u002Fmodule-and-trees`](sub-skills\u002Fmodule-and-trees\u002FSKILL.md)\nUse for `eqx.Module`, field declarations, abstract attributes, PyTree filtering,\n`tree_at`, `tree_equal`, `tree_check`, `apply_updates`, `Partial`, and model\nsurgery.\n\n### [`sub-skills\u002Ffiltered-transformations`](sub-skills\u002Ffiltered-transformations\u002FSKILL.md)\nUse for `filter_jit`, `filter_grad`, `filter_vmap`, `filter_pmap`, `filter_eval_shape`,\n`filter_make_jaxpr`, custom filtered AD, checkpointing, callbacks, and sharding\nbehavior.\n\n### [`sub-skills\u002Fnn-and-state`](sub-skills\u002Fnn-and-state\u002FSKILL.md)\nUse for `equinox.nn` layers, `Sequential`, `Lambda`, `MLP`, `BatchNorm`,\n`State`, `StateIndex`, `Shared`, and inference-mode workflows.\n\n### [`sub-skills\u002Fdiagnostics-and-serialization`](sub-skills\u002Fdiagnostics-and-serialization\u002FSKILL.md)\nUse for `tree_serialise_leaves`, `tree_deserialise_leaves`, runtime errors,\ndebug tools, pretty printing, enumerations, progress meters, and package\ntroubleshooting.\n\n### [`sub-skills\u002Finternal-advanced`](sub-skills\u002Finternal-advanced\u002FSKILL.md)\nUse only for advanced or downstream-library work with `equinox.internal`,\nincluding `noinline`, `while_loop`, `scan`, `nontraceable`, `finalise_jaxpr`,\nand primitive authoring helpers.\n\n## How to choose\n\n- If the task is about model shape, PyTree composition, or field semantics,\n  start with `module-and-trees`.\n- If the task is about JAX transforms crossing model boundaries, start with\n  `filtered-transformations`.\n- If the task is about built-in layers, training loops, or stateful modules,\n  start with `nn-and-state`.\n- If the task is about debugging, serialization, or runtime checks, start with\n  `diagnostics-and-serialization`.\n- If the task explicitly names `equinox.internal` or an advanced helper that is\n  not part of the everyday public API, start with `internal-advanced`.\n\n## Reading order\n\n1. Read the sub-skill that matches the dominant workflow.\n2. Use the linked reference file in that sub-skill for API details, examples,\n   and troubleshooting.\n3. Use [`references\u002Fapi-reference.md`](references\u002Fapi-reference.md) when you need a cross-skill API index.\n4. Use [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md) when the issue looks like install\u002Fimport,\n   backend, optional dependency, or mixed-leaf JAX behavior. Read [`references\u002Frepo-provenance.md`](references\u002Frepo-provenance.md)\n   before relying on version-sensitive guidance.\n\n## Common signals\n\n- A plain `jax.jit` or `jax.grad` call fails because a model contains non-array\n  leaves: route to `filtered-transformations`.\n- A module definition fails because fields or abstract attributes are missing:\n  route to `module-and-trees`.\n- A layer needs `state` or `axis_name`, or `Sequential` has a stateful layer:\n  route to `nn-and-state`.\n- A runtime error, serialization round-trip, or debug trace is the focus:\n  route to `diagnostics-and-serialization`.\n- A user asks about `noinline`, `while_loop`, `scan`, or other semi-public\n  helpers: route to `internal-advanced`.\n\n## Notes\n\n- Equinox is intentionally PyTree-first: models, helper wrappers, and advanced\n  utilities all depend on ordinary JAX semantics rather than a separate runtime.\n- There is no repo CLI. Treat library installation, import checks, and API\n  inspection as the primary entry points.\n- Prefer the bundled references over source checkout links so the skill stays\n  usable after the repository is no longer present.\n","# Equinox\n\nEquinox is a JAX library for building models as PyTrees, applying filtered JAX\ntransformations, and using reusable neural-network layers, stateful modules,\nand advanced debugging utilities.\n\nUse this skill when the user mentions:\n\n- `equinox`, `eqx`, `eqx.Module`, or `equinox.nn`\n- filtered transforms such as `filter_jit`, `filter_grad`, `filter_vmap`, or\n  `filter_pmap`\n- model surgery, PyTree partitioning, serialization, or runtime errors\n- `Sequential`, `MLP`, `BatchNorm`, `State`, `Shared`, or `inference_mode`\n- `equinox.internal` helpers such as `noinline`, `while_loop`, or `scan`\n\n## Quick start\n\nInstall the public package with:\n\n```bash\npip install equinox\n```\n\nFor a local checkout or a health check, run the bundled [`scripts\u002Fsmoke.py`](scripts\u002Fsmoke.py)\nfrom the generated skill root (the directory containing this `SKILL.md`). Set\n`EQUINOX_SKILL_ROOT` to that absolute directory and keep it as the command\nworking directory:\n\n```bash\ncd \"$EQUINOX_SKILL_ROOT\" && python scripts\u002Fsmoke.py --mode all\n```\n\nIf you need multi-device CPU coverage for `pmap` or sharding checks, add\n`--two-cpu-devices` to the smoke script.\n\n## Route map\n\n### [`sub-skills\u002Fmodule-and-trees`](sub-skills\u002Fmodule-and-trees\u002FSKILL.md)\nUse for `eqx.Module`, field declarations, abstract attributes, PyTree filtering,\n`tree_at`, `tree_equal`, `tree_check`, `apply_updates`, `Partial`, and model\nsurgery.\n\n### [`sub-skills\u002Ffiltered-transformations`](sub-skills\u002Ffiltered-transformations\u002FSKILL.md)\nUse for `filter_jit`, `filter_grad`, `filter_vmap`, `filter_pmap`, `filter_eval_shape`,\n`filter_make_jaxpr`, custom filtered AD, checkpointing, callbacks, and sharding\nbehavior.\n\n### [`sub-skills\u002Fnn-and-state`](sub-skills\u002Fnn-and-state\u002FSKILL.md)\nUse for `equinox.nn` layers, `Sequential`, `Lambda`, `MLP`, `BatchNorm`,\n`State`, `StateIndex`, `Shared`, and inference-mode workflows.\n\n### [`sub-skills\u002Fdiagnostics-and-serialization`](sub-skills\u002Fdiagnostics-and-serialization\u002FSKILL.md)\nUse for `tree_serialise_leaves`, `tree_deserialise_leaves`, runtime errors,\ndebug tools, pretty printing, enumerations, progress meters, and package\ntroubleshooting.\n\n### [`sub-skills\u002Finternal-advanced`](sub-skills\u002Finternal-advanced\u002FSKILL.md)\nUse only for advanced or downstream-library work with `equinox.internal`,\nincluding `noinline`, `while_loop`, `scan`, `nontraceable`, `finalise_jaxpr`,\nand primitive authoring helpers.\n\n## How to choose\n\n- If the task is about model shape, PyTree composition, or field semantics,\n  start with `module-and-trees`.\n- If the task is about JAX transforms crossing model boundaries, start with\n  `filtered-transformations`.\n- If the task is about built-in layers, training loops, or stateful modules,\n  start with `nn-and-state`.\n- If the task is about debugging, serialization, or runtime checks, start with\n  `diagnostics-and-serialization`.\n- If the task explicitly names `equinox.internal` or an advanced helper that is\n  not part of the everyday public API, start with `internal-advanced`.\n\n## Reading order\n\n1. Read the sub-skill that matches the dominant workflow.\n2. Use the linked reference file in that sub-skill for API details, examples,\n   and troubleshooting.\n3. Use [`references\u002Fapi-reference.md`](references\u002Fapi-reference.md) when you need a cross-skill API index.\n4. Use [`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md) when the issue looks like install\u002Fimport,\n   backend, optional dependency, or mixed-leaf JAX behavior. Read [`references\u002Frepo-provenance.md`](references\u002Frepo-provenance.md)\n   before relying on version-sensitive guidance.\n\n## Common signals\n\n- A plain `jax.jit` or `jax.grad` call fails because a model contains non-array\n  leaves: route to `filtered-transformations`.\n- A module definition fails because fields or abstract attributes are missing:\n  route to `module-and-trees`.\n- A layer needs `state` or `axis_name`, or `Sequential` has a stateful layer:\n  route to `nn-and-state`.\n- A runtime error, serialization round-trip, or debug trace is the focus:\n  route to `diagnostics-and-serialization`.\n- A user asks about `noinline`, `while_loop`, `scan`, or other semi-public\n  helpers: route to `internal-advanced`.\n\n## Notes\n\n- Equinox is intentionally PyTree-first: models, helper wrappers, and advanced\n  utilities all depend on ordinary JAX semantics rather than a separate runtime.\n- There is no repo CLI. Treat library installation, import checks, and API\n  inspection as the primary entry points.\n- Prefer the bundled references over source checkout links so the skill stays\n  usable after the repository is no longer present.\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\u002Fequinox",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 equinox",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill equinox",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill equinox","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fequinox\u002FSKILL.md",[47,53,57,61,65],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fapi-reference.md","REFERENCE","参考文档",7576,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fequinox\u002Freferences\u002Fapi-reference.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Frepo-provenance.md",1767,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fequinox\u002Freferences\u002Frepo-provenance.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Frepo-routing-metadata.json",442,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fequinox\u002Freferences\u002Frepo-routing-metadata.json",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Ftroubleshooting.md",6955,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fequinox\u002Freferences\u002Ftroubleshooting.md",{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Fsmoke.py","SCRIPT","脚本",7831,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fequinox\u002Fscripts\u002Fsmoke.py",[72,79,86,93,100,107],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"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.","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":14,"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":14,"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":14,"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":14,"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":14,"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",[],[]]