[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75M1RK003ZKT5MVQ4R2CSYV":3,"skill-same:01M75M1RK003ZKT5MVQ4R2CSYV":95},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M75M1RK003ZKT5MVQ4R2CSYV","lean-project-architecture",null,"Use when planning a Lean formalization, selecting and freezing author-approved declarations, choosing module boundaries, designing reusable APIs, or staging a large proof project into stable phases.","stanfish06\u002Fskillquarium","stanfish06","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F93099613?v=4","01M75F1CK4ZW109ZK3T26VN52F","Zhiyuan","01M75F1CM53VH6F79DZ8GXKKES",12,"CC-BY-4.0","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: lean-project-architecture\ndescription: Use when planning a Lean formalization, selecting and freezing author-approved declarations, choosing module boundaries, designing reusable APIs, or staging a large proof project into stable phases.\n---\n\n# Lean project architecture\n\nUse this skill when the main task is project design rather than one proof.\nTurn the mathematical program into stable source-facing Lean declarations,\nordinary implementation modules, and a dependency graph that can be built\nincrementally. Repository rules override this skill.\n\n## The constitutional surface is actual Lean code\n\nSelect a small surface: main roots, subsection or induction gates, and\ndefinitions whose meaning changes them. Do not freeze every helper lemma.\n\nOnly an actual final Lean declaration can be approved, manifested, frozen, or\nused as a dependency-graph anchor. Markdown “contracts,” candidate APIs,\ntracked probes, placeholder declarations, temporary carriers, fallback bodies,\nand promises of later characterization are planning debris, not mathematics.\nPre-approval elaboration may use an untracked temporary file containing the\nexact intended final command; delete it after use.\n\nFor each source-facing declaration:\n\n1. pin the mathematical source, errata, and author rulings;\n2. transcribe its premises, conclusion, quantifier order, constant\u002Fwitness\n   scope, and split\u002Fjoint policy;\n3. identify and recursively inspect every meaning-carrying carrier, definition,\n   typeclass, and bundled assumption;\n4. elaborate the exact final theorem type or exact final definition body;\n5. obtain independent source\u002Fbinder\u002Fsemantics review;\n6. show the complete Lean command to the author; and\n7. only after approval, put one declaration in one frozen file and bind its\n   source, path, fully qualified export, frozen bytes, and state in a manifest.\n\nA topic list or source transcription is useful planning evidence but has no\ndeclaration status. Old Lean code and old graphs are evidence to mine only\nafter the source-facing code is fixed.\n\n## Freeze keystone theorem statements\n\nAn exact author-approved keystone proposition or theorem should initially have\nthe manifest-bound proof body `by sorry` and state `DRAFT_SORRY`. Freeze it\nbefore graph extraction and proof assignment, rather than allowing its type to\ntrack proof convenience. Its type, binders, imports, and meaning-carrying\ndefinitions are already final. It is `FROZEN_UNPROVED`, not proof progress, and\nproduction\u002Fprovider modules may not import or consume it.\n\nAllow this only in one-declaration frozen files with manifest identity, source\nand declaration hashes, a named provider, and enforced quarantine. All other\n`sorry`, all axioms, and every provisional definition remain forbidden.\n\nDefinition types and bodies are never provisional and neither contain nor\ndepend on a draft placeholder. A genuine manuscript-defined predicate may be\na literal `def : Prop`; a theorem-shaped predicate, assumption package, result\nbundle, or generic “input\u002Fbridge\u002Fconsequence” proposition may not replace a\ntheorem.\n\n## Source-characterized definitions\n\nA definition is acceptable only when its body already denotes the manuscript\nobject on its whole public carrier. Do not use an over-broad carrier plus an\n`if`\u002F`dite`\u002Fzero\u002Fdefault branch, choose an arbitrary non-unique witness, or\nland a reconstruction whose source identity remains a future theorem.\n\nIf the source defines an object as the unique object satisfying `Φ`:\n\n1. freeze the exact well-definedness theorem `∃! x, Φ x` as a theorem goal;\n2. prove, seal, and independently audit it;\n3. define the object from that proved theorem; and\n4. land the definition together with a separately named, sorry-free theorem\n   asserting its full source characterization and uniqueness.\n\nThe well-definedness theorem may not assume response existence, integrability,\nlinearity, quadraticity, representability, or another proof step unless the\nsource states it as a premise. Those are dependencies to prove and apply. A\nClassical.choose from the proved ∃! is permitted; a fallback or non-unique\nchoice is not.\n\nTreat the definition and characterization theorem as one semantic unit:\ngive each a manifest entry, land them in the same commit, version and audit\nthem together, and route every downstream graph node through\nthe characterization theorem rather than the raw result type. A literal source\nformula or predicate may instead be defined directly after its body and carrier\nhave passed the same audit.\n\n## Dependency and module architecture\n\n1. Build the source dependency graph against exact frozen exports.\n   - distinguish external black boxes, reusable infrastructure, and\n     paper-specific steps;\n   - make graph edges proof obligations, never added hypotheses;\n   - record alternative routes explicitly instead of conjoining them;\n   - review topology independently.\n2. Put theorem providers below anchors in the import DAG. Providers target\n   separate exports and never import drafts. Until a frozen anchor is sealed\n   and audited, its registered provider module\u002Fexport is itself quarantined\n   from every other consumer.\n3. Keep reusable infrastructure separate from paper-specific assembly.\n4. Put declarations in the earliest focused module where they naturally belong.\n5. Minimize imports in foundational files.\n6. Search Mathlib and approved upstream APIs before inventing definitions.\n7. Build APIs before downstream proofs:\n   - the real definitions;\n   - source characterization and basic access lemmas;\n   - extensionality and simplification lemmas;\n   - stable notation and coercion policy.\n8. Split work into phases with stop\u002Fgo checkpoints. Seal a theorem by changing\n   only its proof body to a short `by exact Provider.result`, then run exact\n   statement, axiom, and source audits.\n\n## Lean module system (when supported or required)\n\nWhen the selected Lean toolchain and project policy use the module system,\ndesign every regular `.lean` file as a module from the start. A module cannot\nimport a non-module file, so mixed dependency trees must be ported bottom-up.\nDo not impose this layout on a toolchain or established project that does not\nsupport or require it.\n\n- **File shape.** Copyright comment; `module`; imports; module docstring;\n  `public section` (or `@[expose] public section`).\n- **Imports.** `public import` whatever the public interface needs; plain\n  `import` for proof-only dependencies.\n- **Exposure.** Definition bodies are hidden unless `@[expose]`, even from\n  exported theorems in the same file. Expose source-facing, frozen and\n  Challenge definitions.\n- **Dependencies.** Check at planning time that every dependency is modular.\n  Port bottom-up, before freezing statements when the order is still open.\n- **Size.** Follow the project's limit. Registry submissions may impose a\n  10,000-line cap.\n- **Checks.** Run [scripts\u002Fcheck_lean_modules.py](scripts\u002Fcheck_lean_modules.py)\n  `--root \u003Crepo>` when the repository adopts this module\u002Fregistry profile\n  (stage new files first; it reads the git index). It enforces the module\n  header, configured line cap, and no symlinked `.lean` files. If the project\n  contains `comparator.json`, it also applies the bundled registry checks.\n  Its lexer helpers are the model for a project's own checkers.\n\nRead [references\u002Fmodule-system.md](references\u002Fmodule-system.md) before writing\nor porting a repository that uses Lean modules, and its \"Checker scripts\"\nsection before writing or porting any script that reads modular Lean source.\n\n## Statement-integrity rules\n\n- A source-facing theorem has exactly source premises, standing assumptions,\n  typing data, and explicit author rulings. Proof obligations stay in proofs.\n- Expand project-owned structures and typeclasses during audit; one bundled\n  argument can hide dozens of excess hypotheses.\n- Audit binder order for uniform constants. An existential `C` must precede\n  every parameter on which the source says it is independent.\n- If one source lemma is split, require explicit author approval. Preserve a\n  shared constant or witness in one joint declaration unless independent\n  choices are expressly allowed.\n- Conditional helpers are ordinary internal lemmas with distinct names. They\n  never certify or replace the source-facing theorem.\n- Frozen declarations are immutable. A defect requires an author-approved\n  versioned successor and invalidates dependent graph\u002Faudit evidence; never add\n  a premise or silently substitute a helper.\n\n## Performance and maintenance\n\n- Keep files focused and within repository size limits.\n- Prefer small helper lemmas over monolithic proofs.\n- Obey repository heartbeat policy; a default-heartbeat failure is an\n  architectural signal when overrides are forbidden.\n- Profile before changing architecture or adding typeclass caches.\n- Treat linter cleanliness and build stability as design concerns.\n- Preserve other sessions' files and follow project-specific build, cache,\n  staging, and landing rules.\n\n## References\n\nRead [references\u002Fproject-architecture.md](references\u002Fproject-architecture.md)\nfor definition engineering, file organization, Mathlib-facing conventions,\nperformance guidance, and cleanup expectations. Read\n[references\u002Fmodule-system.md](references\u002Fmodule-system.md) for module-system\nfile shape, visibility, porting, and conditional registry limits.\n","# Lean project architecture\n\nUse this skill when the main task is project design rather than one proof.\nTurn the mathematical program into stable source-facing Lean declarations,\nordinary implementation modules, and a dependency graph that can be built\nincrementally. Repository rules override this skill.\n\n## The constitutional surface is actual Lean code\n\nSelect a small surface: main roots, subsection or induction gates, and\ndefinitions whose meaning changes them. Do not freeze every helper lemma.\n\nOnly an actual final Lean declaration can be approved, manifested, frozen, or\nused as a dependency-graph anchor. Markdown “contracts,” candidate APIs,\ntracked probes, placeholder declarations, temporary carriers, fallback bodies,\nand promises of later characterization are planning debris, not mathematics.\nPre-approval elaboration may use an untracked temporary file containing the\nexact intended final command; delete it after use.\n\nFor each source-facing declaration:\n\n1. pin the mathematical source, errata, and author rulings;\n2. transcribe its premises, conclusion, quantifier order, constant\u002Fwitness\n   scope, and split\u002Fjoint policy;\n3. identify and recursively inspect every meaning-carrying carrier, definition,\n   typeclass, and bundled assumption;\n4. elaborate the exact final theorem type or exact final definition body;\n5. obtain independent source\u002Fbinder\u002Fsemantics review;\n6. show the complete Lean command to the author; and\n7. only after approval, put one declaration in one frozen file and bind its\n   source, path, fully qualified export, frozen bytes, and state in a manifest.\n\nA topic list or source transcription is useful planning evidence but has no\ndeclaration status. Old Lean code and old graphs are evidence to mine only\nafter the source-facing code is fixed.\n\n## Freeze keystone theorem statements\n\nAn exact author-approved keystone proposition or theorem should initially have\nthe manifest-bound proof body `by sorry` and state `DRAFT_SORRY`. Freeze it\nbefore graph extraction and proof assignment, rather than allowing its type to\ntrack proof convenience. Its type, binders, imports, and meaning-carrying\ndefinitions are already final. It is `FROZEN_UNPROVED`, not proof progress, and\nproduction\u002Fprovider modules may not import or consume it.\n\nAllow this only in one-declaration frozen files with manifest identity, source\nand declaration hashes, a named provider, and enforced quarantine. All other\n`sorry`, all axioms, and every provisional definition remain forbidden.\n\nDefinition types and bodies are never provisional and neither contain nor\ndepend on a draft placeholder. A genuine manuscript-defined predicate may be\na literal `def : Prop`; a theorem-shaped predicate, assumption package, result\nbundle, or generic “input\u002Fbridge\u002Fconsequence” proposition may not replace a\ntheorem.\n\n## Source-characterized definitions\n\nA definition is acceptable only when its body already denotes the manuscript\nobject on its whole public carrier. Do not use an over-broad carrier plus an\n`if`\u002F`dite`\u002Fzero\u002Fdefault branch, choose an arbitrary non-unique witness, or\nland a reconstruction whose source identity remains a future theorem.\n\nIf the source defines an object as the unique object satisfying `Φ`:\n\n1. freeze the exact well-definedness theorem `∃! x, Φ x` as a theorem goal;\n2. prove, seal, and independently audit it;\n3. define the object from that proved theorem; and\n4. land the definition together with a separately named, sorry-free theorem\n   asserting its full source characterization and uniqueness.\n\nThe well-definedness theorem may not assume response existence, integrability,\nlinearity, quadraticity, representability, or another proof step unless the\nsource states it as a premise. Those are dependencies to prove and apply. A\nClassical.choose from the proved ∃! is permitted; a fallback or non-unique\nchoice is not.\n\nTreat the definition and characterization theorem as one semantic unit:\ngive each a manifest entry, land them in the same commit, version and audit\nthem together, and route every downstream graph node through\nthe characterization theorem rather than the raw result type. A literal source\nformula or predicate may instead be defined directly after its body and carrier\nhave passed the same audit.\n\n## Dependency and module architecture\n\n1. Build the source dependency graph against exact frozen exports.\n   - distinguish external black boxes, reusable infrastructure, and\n     paper-specific steps;\n   - make graph edges proof obligations, never added hypotheses;\n   - record alternative routes explicitly instead of conjoining them;\n   - review topology independently.\n2. Put theorem providers below anchors in the import DAG. Providers target\n   separate exports and never import drafts. Until a frozen anchor is sealed\n   and audited, its registered provider module\u002Fexport is itself quarantined\n   from every other consumer.\n3. Keep reusable infrastructure separate from paper-specific assembly.\n4. Put declarations in the earliest focused module where they naturally belong.\n5. Minimize imports in foundational files.\n6. Search Mathlib and approved upstream APIs before inventing definitions.\n7. Build APIs before downstream proofs:\n   - the real definitions;\n   - source characterization and basic access lemmas;\n   - extensionality and simplification lemmas;\n   - stable notation and coercion policy.\n8. Split work into phases with stop\u002Fgo checkpoints. Seal a theorem by changing\n   only its proof body to a short `by exact Provider.result`, then run exact\n   statement, axiom, and source audits.\n\n## Lean module system (when supported or required)\n\nWhen the selected Lean toolchain and project policy use the module system,\ndesign every regular `.lean` file as a module from the start. A module cannot\nimport a non-module file, so mixed dependency trees must be ported bottom-up.\nDo not impose this layout on a toolchain or established project that does not\nsupport or require it.\n\n- **File shape.** Copyright comment; `module`; imports; module docstring;\n  `public section` (or `@[expose] public section`).\n- **Imports.** `public import` whatever the public interface needs; plain\n  `import` for proof-only dependencies.\n- **Exposure.** Definition bodies are hidden unless `@[expose]`, even from\n  exported theorems in the same file. Expose source-facing, frozen and\n  Challenge definitions.\n- **Dependencies.** Check at planning time that every dependency is modular.\n  Port bottom-up, before freezing statements when the order is still open.\n- **Size.** Follow the project's limit. Registry submissions may impose a\n  10,000-line cap.\n- **Checks.** Run [scripts\u002Fcheck_lean_modules.py](scripts\u002Fcheck_lean_modules.py)\n  `--root \u003Crepo>` when the repository adopts this module\u002Fregistry profile\n  (stage new files first; it reads the git index). It enforces the module\n  header, configured line cap, and no symlinked `.lean` files. If the project\n  contains `comparator.json`, it also applies the bundled registry checks.\n  Its lexer helpers are the model for a project's own checkers.\n\nRead [references\u002Fmodule-system.md](references\u002Fmodule-system.md) before writing\nor porting a repository that uses Lean modules, and its \"Checker scripts\"\nsection before writing or porting any script that reads modular Lean source.\n\n## Statement-integrity rules\n\n- A source-facing theorem has exactly source premises, standing assumptions,\n  typing data, and explicit author rulings. Proof obligations stay in proofs.\n- Expand project-owned structures and typeclasses during audit; one bundled\n  argument can hide dozens of excess hypotheses.\n- Audit binder order for uniform constants. An existential `C` must precede\n  every parameter on which the source says it is independent.\n- If one source lemma is split, require explicit author approval. Preserve a\n  shared constant or witness in one joint declaration unless independent\n  choices are expressly allowed.\n- Conditional helpers are ordinary internal lemmas with distinct names. They\n  never certify or replace the source-facing theorem.\n- Frozen declarations are immutable. A defect requires an author-approved\n  versioned successor and invalidates dependent graph\u002Faudit evidence; never add\n  a premise or silently substitute a helper.\n\n## Performance and maintenance\n\n- Keep files focused and within repository size limits.\n- Prefer small helper lemmas over monolithic proofs.\n- Obey repository heartbeat policy; a default-heartbeat failure is an\n  architectural signal when overrides are forbidden.\n- Profile before changing architecture or adding typeclass caches.\n- Treat linter cleanliness and build stability as design concerns.\n- Preserve other sessions' files and follow project-specific build, cache,\n  staging, and landing rules.\n\n## References\n\nRead [references\u002Fproject-architecture.md](references\u002Fproject-architecture.md)\nfor definition engineering, file organization, Mathlib-facing conventions,\nperformance guidance, and cleanup expectations. Read\n[references\u002Fmodule-system.md](references\u002Fmodule-system.md) for module-system\nfile shape, visibility, porting, and conditional registry limits.\n",{"name":5,"description":7},"master","49e49abf2f670d429e0b7e2c0349739c3fdd5cbe","skills\u002Flean-project-architecture",5,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium","https:\u002F\u002Fgithub.com\u002Fstanfish06","https:\u002F\u002Fwww.stanfish.dev\u002F","2026-11-14 21:31:37",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add stanfish06\u002Fskillquarium --skill lean-project-architecture",{"tool":39,"command":40},"bunx","bunx skills add stanfish06\u002Fskillquarium --skill lean-project-architecture",{"tool":42,"command":43},"pnpm","pnpm dlx skills add stanfish06\u002Fskillquarium --skill lean-project-architecture","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FSKILL.md",true,[47,53,57,61,65,69,73,77,83,87,91],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"agents\u002Fopenai.yaml","REFERENCE","参考文档",312,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002Fagents\u002Fopenai.yaml",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"INDEX.md",2981,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FINDEX.md",{"path":58,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":16},"LICENSE","许可证",18656,{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"LICENSES\u002FApache-2.0.txt",11358,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FLICENSES\u002FApache-2.0.txt",{"path":66,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":67,"sourceHtmlUrl":68},"PROVENANCE.md",2674,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FPROVENANCE.md",{"path":70,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Fmodule-system.md",8673,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002Freferences\u002Fmodule-system.md",{"path":74,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fproject-architecture.md",8519,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002Freferences\u002Fproject-architecture.md",{"path":78,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":81,"sourceHtmlUrl":82},"scripts\u002Fcheck_lean_modules.py","SCRIPT","脚本",14244,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002Fscripts\u002Fcheck_lean_modules.py",{"path":84,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":85,"sourceHtmlUrl":86},"SOURCES.md",1513,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FSOURCES.md",{"path":88,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":89,"sourceHtmlUrl":90},"tests\u002Ftest_check_lean_modules.py",11500,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002Ftests\u002Ftest_check_lean_modules.py",{"path":92,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":93,"sourceHtmlUrl":94},"THIRD_PARTY_NOTICES.md",2165,"https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Flean-project-architecture\u002FTHIRD_PARTY_NOTICES.md",[96,104,113,120,127,134],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":100,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":20},"01M76ZJTRVA456GD7K8Y3EP82H","zarr-python","Stores and queries chunked N-D scientific arrays with Zarr-Python 3, including codecs, sharding, S3\u002FGCS storage, and NumPy\u002FDask\u002FXarray integration. Use for array layout, bounded I\u002FO, format migration, or scientific metadata preservation.","MIT","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fzarr-python\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":109,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":20},"01M76ZJQZ15Q1HKC5V85NHE325","yeet","Publish local changes to GitHub by confirming scope, committing intentionally, pushing the branch, and opening a draft PR through the GitHub app from this plugin, with `gh` used only as a fallback where connector coverage is insufficient.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fyeet\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fyeet\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":20},"01M76ZJ8ASZ4QYQ70Y6SBA4Z4S","xarray","N-dimensional labeled arrays and datasets in Python. Built on top of NumPy and Dask. It introduces labels in the form of dimensions, coordinates, and attributes on top of raw NumPy-like arrays, making data analysis in physical sciences more intuitive and less error-prone. Use for working with multi-dimensional scientific data, NetCDF\u002FGRIB\u002FZarr files, climate\u002Fweather\u002Foceanographic datasets, remote sensing, geospatial imaging, large out-of-memory datasets with Dask, and labeled array operations.","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fxarray\u002FSKILL.md",[],[],{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":100,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":126,"publishedAt":20},"01M76ZHFBXD6VXPQ02VCEG202B","writing-mstest-tests","Use when asked to fix, rewrite, update, improve, modernize, show corrected code for, or explain existing MSTest tests or MSTest-specific configuration. Use for \"review\" when corrected code or edits are wanted, even for one pasted assertion or passing tests with bad failure output. Covers expected\u002Factual labels; generic Boolean, collection, string, numeric, null, identity, exception, hard-cast, and object[] checks; TestContext\u002Flifecycle; timeout\u002Fcancellation; OS\u002FCI conditions, retry, cleanup, parallelization, MSTest.Sdk project setup, and MSTESTxxxx. Honor the installed MSTest version. DO NOT USE to design new test cases (code-testing), perform report-only audits, create project files rather than explain MSTest setup, run tests, migrate frameworks, or handle non-MSTest\u002Fnon-.NET code.","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fwriting-mstest-tests\u002FSKILL.md",[],[],{"skillNo":128,"name":129,"nameZh":6,"description":130,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":100,"licenseFileUrl":6,"sourceHtmlUrl":131,"categories":132,"keywords":133,"publishedAt":20},"01M76ZDFDF4N88RH6HSYGS7GP2","wgs-prs","End-to-end WGS to polygenic risk score pipeline. Takes paired-end FASTQ files (or a pre-existing VCF) through nf-core\u002Fsarek for variant calling, applies VCF QC (normalisation, hard filtering, Ti\u002FTv and Het\u002FHom checks), then computes polygenic risk scores via the PGS Catalog. Fills the FASTQ to VCF gap upstream of the gwas-prs skill.","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fwgs-prs\u002FSKILL.md",[],[],{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":138,"sourceHtmlUrl":139,"categories":140,"keywords":141,"publishedAt":20},"01M76ZCC1SC68J42QK4MDC8RCG","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fstanfish06\u002Fskillquarium\u002Fblob\u002Fmaster\u002Fskills\u002Fwebapp-testing\u002FSKILL.md",[],[]]