[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79XNGMTSJG2RM226C85A719":3,"skill-same:01M79XNGMTSJG2RM226C85A719":47},{"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":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M79XNGMTSJG2RM226C85A719","deniz-dotnet-general.vectorization",null,"Design, implement, optimize, and review SIMD code in .NET. USE FOR: vectorizing scalar loops with TensorPrimitives, Vector64\u002F128\u002F256\u002F512, or platform hardware intrinsics; reviewing existing SIMD code, including the generic Vector type, for contract equivalence, tail handling, memory safety, portability, fallbacks, and measured performance. DO NOT USE FOR: performance work unrelated to SIMD or vectorization.","Blind-Striker\u002Fagent-skills-and-plugins","Blind-Striker","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1965259?v=4","01M79WFXTE5PH0C1VB7B54ANMN","Deniz İrgin","01M79WFXV5KZ5GST1K8G035DM5",1,"MIT","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fopencode\u002Fdeniz-dotnet-general\u002Fskills\u002Fdeniz-dotnet-general.vectorization\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: deniz-dotnet-general.vectorization\ndescription: >\n  Design, implement, optimize, and review SIMD code in .NET. USE FOR:\n  vectorizing scalar loops with TensorPrimitives, Vector64\u002F128\u002F256\u002F512, or\n  platform hardware intrinsics; reviewing existing SIMD code, including the\n  generic Vector type, for contract equivalence, tail handling, memory safety,\n  portability, fallbacks, and measured performance. DO NOT USE FOR: performance\n  work unrelated to SIMD or vectorization.\n---\n\n# .NET SIMD vectorization\n\nProduce a portable optimization that preserves the scalar contract, remains memory-safe at every\nlength, and earns its complexity with measured results. **Read the official\n[SIMD and hardware-intrinsics guidance](https:\u002F\u002Flearn.microsoft.com\u002Fdotnet\u002Fstandard\u002Fsimd) first**\nand follow its comprehensive implementation templates. In particular, use its self-contained\nper-width dispatch, dedicated small-input handling, loop, and remainder shapes rather than reducing\nthem to a chain of width checks. This skill supplies the decision rules and validation checks to\napply while changing real code.\n\n## Inputs and prerequisites\n\nDiscover these from the repository before asking the user:\n\n| Input | Required | What to establish |\n| --- | --- | --- |\n| Scalar implementation and tests | Yes | Existing contract, representative call sites, and supported overlap |\n| Target frameworks and platforms | Yes | Available SIMD APIs and architectures that must behave consistently |\n| Build and test workflow | Yes | The repository's normal commands and how to launch separate test processes |\n| Representative workload or benchmark | For optimization | Typical input sizes and the baseline to beat |\n\nDo not add a package merely because an API exists there. First check the target framework and the\nproject's existing dependency\u002Fversioning policy.\n\n## Core rules\n\n1. **Use the highest-level API that matches the contract, then stop.** `Span\u003CT>` and `string`\n   operations, `TensorPrimitives`, and tensor types already accelerate many operations. LINQ\n   reductions such as `Sum`, `Min`, `Max`, and `Average` can also accelerate when the source exposes\n   its underlying span. Verify empty-input and floating-point behavior rather than assuming similarly\n   named operations are interchangeable. Once an existing API preserves the contract, use it instead\n   of continuing into handwritten SIMD. Before writing an explicit loop, name the framework APIs\n   considered and why none applies. Fixed-shape `System.Numerics` types remain appropriate for\n   graphics and similar domains.\n2. **Start new explicit SIMD loops with `Vector128\u003CT>`.** It is accelerated across the broadest\n   hardware set. Add wider fixed-width paths only when measurements justify them.\n3. **Keep platforms consistent.** Prefer cross-platform operations on the fixed-width vector types;\n   they lower to the appropriate target instructions. For example,\n   `(vector & mask) == Vector128\u003Cbyte>.Zero` becomes `ptest` on x86\u002Fx64. Use\n   architecture-specific intrinsics only for a measured gap, guard them with `IsSupported`, and\n   retain equivalent portable or scalar behavior.\n4. **Read `IsHardwareAccelerated`, `IsSupported`, and `Count` directly.** The JIT treats them as\n   constants, so caching them adds no value and obscures which branches disappear.\n5. **Prefer operators where they are clear.** Parenthesize expressions that mix bitwise and\n   comparison operators so precedence is explicit.\n\nIf the task is review-only, do not rewrite the code. Report correctness and memory-safety defects\nbefore performance opportunities.\n\n## Authoring checklist\n\n- **Contract:** identify behavior for empty and short inputs, overlap, overflow, NaN, signed zero,\n  ordering, and exceptions before changing the implementation.\n- **Framework gate:** inspect the target framework and existing package references, then compile or\n  probe the highest-level candidate API with the required edge cases. A small contract adapter, such\n  as preserving special empty-input behavior, does not justify reimplementing the operation. If the\n  API preserves the contract, use it and stop; do not claim it is unavailable without checking.\n- **Structure:** for new explicit SIMD, implement `Vector128\u003CT>` and scalar first. Only after\n  measurements justify wider paths, check `Vector512\u003CT>`, then `Vector256\u003CT>`, optional `Vector\u003CT>`,\n  `Vector128\u003CT>`, and finally scalar. Omit paths the implementation does not need. Each outer\n  fixed-width guard checks only its `IsHardwareAccelerated` property and, for generic element types,\n  `IsSupported`. Inside that block, run the width-specific helper when the input has at least\n  `Count` elements; otherwise run a dedicated small-input helper, then return. Do not put the length\n  check in the outer guard and fall through to repeat dispatch at narrower widths. Keeping each\n  supported-width block self-contained lets the JIT remove unsupported blocks and avoids redundant\n  work on common small inputs.\n- **Loads and stores:** prefer span-based `Vector128.Create(span)` and `CopyTo`; the JIT keeps them\n  efficient and they require no pinning or reference arithmetic. Unsafe loads and stores are largely\n  unnecessary. When a path genuinely must walk a buffer by managed reference, use the element-offset\n  `LoadUnsafe(ref T, nuint)` and `StoreUnsafe` overloads rather than pointers or manually advanced\n  references.\n- **Empty inputs:** in a reference-based path, obtain the starting reference with\n  `MemoryMarshal.GetReference(span)` or `MemoryMarshal.GetArrayDataReference(array)`, not by indexing\n  element `0`.\n- **Unsupported element types:** the fixed-width vectors support primitive numeric element types,\n  not `char` or `bool`. Reinterpret with `MemoryMarshal.Cast` or `As\u003CTFrom, TTo>`; reinterpretation\n  changes only the type, not the bits. Keep Boolean data as `0` or `1` and characters as valid\n  UTF-16, normalizing results before storing when necessary.\n- **Offsets:** prove the input contains a full vector before subtracting `Count` or converting an\n  index to `nuint`; otherwise a negative value becomes a huge unsigned offset.\n- **Managed references:** do not form references before the start or past the end of a span,\n  including a one-past-end reference. The runtime permits a non-dereferenced managed pointer exactly\n  one past an object or array, but this guidance intentionally prohibits the pattern because it is\n  fragile and easy to misuse. Keep the base reference in range and express traversal with an element\n  offset.\n- **Remainders:** cover every length, including `0`, `Count - 1`, `Count`, `Count + 1`, and\n  nonmultiples of each width. Once the input contains a full vector, keep the tail vectorized by\n  reprocessing the last full vector. An idempotent operation can fold that overlap in directly. A\n  non-idempotent operation must use `ConditionalSelect` to replace repeated lanes with the\n  operation's identity before folding them in. This is the JIT-recognized general pattern; it can\n  reduce a zero-identity selection to a bitwise mask while retaining broader optimization\n  opportunities. For in-place transforms, preserve the original tail values before overlapping\n  stores and write only valid results.\n- **Buffer overlap:** choose a traversal direction or staging strategy that prevents stores from\n  corrupting values not yet loaded.\n- **Numeric behavior:** account for floating-point reassociation, NaN and signed-zero semantics,\n  checked or unchecked integer overflow, and endianness where the algorithm depends on byte order.\n  `Native` and `Estimate` operations can intentionally relax precision or IEEE edge-case behavior;\n  use them only when the contract permits it and measurements justify them.\n\nThe official guidance contains the complete dispatch, small-input, unrolling, and remainder\ntemplates; use those for the full implementation. The following excerpt illustrates only the inner\nsafe `Vector128\u003CT>` loop for an in-place elementwise transform, after its self-contained dispatch\nblock has established at least one full vector. `Transform` represents the operation being\nimplemented:\n\n```csharp\nSpan\u003Cint> tail = data.Slice(data.Length - Vector128\u003Cint>.Count);\nVector128\u003Cint> end = Vector128.Create\u003Cint>(tail);\nSpan\u003Cint> remaining = data;\n\nwhile (remaining.Length >= Vector128\u003Cint>.Count)\n{\n    Vector128\u003Cint> values = Vector128.Create\u003Cint>(remaining);\n    Transform(values).CopyTo(remaining);\n    remaining = remaining.Slice(Vector128\u003Cint>.Count);\n}\n\nif (!remaining.IsEmpty)\n{\n    Transform(end).CopyTo(tail);\n}\n```\n\nThe early `end` load preserves original values before overlapping stores. For a read-only reduction,\nload the same final span after the main loop and use `ConditionalSelect` to replace already-processed\nlanes with the operation's identity. Do not substitute `LoadUnsafe`\u002F`StoreUnsafe` or a scalar\nepilogue merely to avoid span bounds checks.\n\n## Testing checklist\n\n- Compare the optimized implementation with the scalar contract across boundary lengths,\n  randomized values, empty inputs, supported overlap, and numeric edge cases. Cover every\n  implemented width and the scalar path with inputs both large enough and too small to benefit.\n- Exercise every implemented width and the scalar fallback in separate processes. On x86\u002Fx64\n  CoreCLR, `DOTNET_EnableAVX2=0` disables AVX2 and `DOTNET_EnableHWIntrinsic=0` disables hardware\n  intrinsics. Use the repository's normal test command and do not change these process-wide\n  settings inside a unit test. These settings do not change code already compiled as ReadyToRun or\n  ahead of time, so confirm the target code is JIT-compiled when using them to force a path.\n- For unsafe loads and stores, use guard-page or equivalent boundary tests when available. Put the\n  inaccessible page after the buffer for forward iteration and before it for backwards iteration,\n  and include nonmultiple lengths. An ordinary array allocation does not reliably expose an\n  out-of-bounds read.\n\n## Benchmarking\n\nUse BenchmarkDotNet to measure representative small and large inputs before keeping the added\ncomplexity. Compare scalar, `Vector128\u003CT>`, and each wider implemented path in the same run. Small\ninputs can be slower because setup dominates, and speedups are rarely the theoretical vector-width\nmultiple because memory throughput, alignment, and latency still apply. Report throughput or time\nwith noise context and, when relevant, generated code size or instruction counts. Control allocation\nalignment for stable measurements or randomize it to observe the distribution. A wider vector is\nnot automatically faster.\n\nIf the project cannot target the required framework, run the relevant architecture, or execute the\nfallback configuration, state exactly which path remains unverified. Do not claim success from a\ndefault-hardware test alone.\n\n## Completion contract\n\n- **Authoring:** leave the scalar contract covered by tests; identify the framework or SIMD layer\n  selected; report measurements for the representative workload; name any architecture or fallback\n  path that could not be exercised.\n- **Review:** report only concrete findings, ordered by correctness, memory safety, portability,\n  tests, then performance evidence. If none remain, say so directly.\n- Do not call an optimization complete when it only builds, only passes on the current machine, or\n  has no comparison against the scalar baseline.\n\n## Review checklist\n\nReview in this order:\n\n1. Scalar-contract equivalence, including signed zero, NaN, overflow, and relevant endianness\n2. Reuse of an existing accelerated framework API\n3. Tail correctness for idempotent versus non-idempotent work\n4. Memory safety, unsigned offset arithmetic, empty inputs, and overlapping buffers\n5. Portable dispatch and behaviorally equivalent fallbacks\n6. Tests that force each width and the scalar path\n7. Benchmarks that justify explicit SIMD and additional widths\n","# .NET SIMD vectorization\n\nProduce a portable optimization that preserves the scalar contract, remains memory-safe at every\nlength, and earns its complexity with measured results. **Read the official\n[SIMD and hardware-intrinsics guidance](https:\u002F\u002Flearn.microsoft.com\u002Fdotnet\u002Fstandard\u002Fsimd) first**\nand follow its comprehensive implementation templates. In particular, use its self-contained\nper-width dispatch, dedicated small-input handling, loop, and remainder shapes rather than reducing\nthem to a chain of width checks. This skill supplies the decision rules and validation checks to\napply while changing real code.\n\n## Inputs and prerequisites\n\nDiscover these from the repository before asking the user:\n\n| Input | Required | What to establish |\n| --- | --- | --- |\n| Scalar implementation and tests | Yes | Existing contract, representative call sites, and supported overlap |\n| Target frameworks and platforms | Yes | Available SIMD APIs and architectures that must behave consistently |\n| Build and test workflow | Yes | The repository's normal commands and how to launch separate test processes |\n| Representative workload or benchmark | For optimization | Typical input sizes and the baseline to beat |\n\nDo not add a package merely because an API exists there. First check the target framework and the\nproject's existing dependency\u002Fversioning policy.\n\n## Core rules\n\n1. **Use the highest-level API that matches the contract, then stop.** `Span\u003CT>` and `string`\n   operations, `TensorPrimitives`, and tensor types already accelerate many operations. LINQ\n   reductions such as `Sum`, `Min`, `Max`, and `Average` can also accelerate when the source exposes\n   its underlying span. Verify empty-input and floating-point behavior rather than assuming similarly\n   named operations are interchangeable. Once an existing API preserves the contract, use it instead\n   of continuing into handwritten SIMD. Before writing an explicit loop, name the framework APIs\n   considered and why none applies. Fixed-shape `System.Numerics` types remain appropriate for\n   graphics and similar domains.\n2. **Start new explicit SIMD loops with `Vector128\u003CT>`.** It is accelerated across the broadest\n   hardware set. Add wider fixed-width paths only when measurements justify them.\n3. **Keep platforms consistent.** Prefer cross-platform operations on the fixed-width vector types;\n   they lower to the appropriate target instructions. For example,\n   `(vector & mask) == Vector128\u003Cbyte>.Zero` becomes `ptest` on x86\u002Fx64. Use\n   architecture-specific intrinsics only for a measured gap, guard them with `IsSupported`, and\n   retain equivalent portable or scalar behavior.\n4. **Read `IsHardwareAccelerated`, `IsSupported`, and `Count` directly.** The JIT treats them as\n   constants, so caching them adds no value and obscures which branches disappear.\n5. **Prefer operators where they are clear.** Parenthesize expressions that mix bitwise and\n   comparison operators so precedence is explicit.\n\nIf the task is review-only, do not rewrite the code. Report correctness and memory-safety defects\nbefore performance opportunities.\n\n## Authoring checklist\n\n- **Contract:** identify behavior for empty and short inputs, overlap, overflow, NaN, signed zero,\n  ordering, and exceptions before changing the implementation.\n- **Framework gate:** inspect the target framework and existing package references, then compile or\n  probe the highest-level candidate API with the required edge cases. A small contract adapter, such\n  as preserving special empty-input behavior, does not justify reimplementing the operation. If the\n  API preserves the contract, use it and stop; do not claim it is unavailable without checking.\n- **Structure:** for new explicit SIMD, implement `Vector128\u003CT>` and scalar first. Only after\n  measurements justify wider paths, check `Vector512\u003CT>`, then `Vector256\u003CT>`, optional `Vector\u003CT>`,\n  `Vector128\u003CT>`, and finally scalar. Omit paths the implementation does not need. Each outer\n  fixed-width guard checks only its `IsHardwareAccelerated` property and, for generic element types,\n  `IsSupported`. Inside that block, run the width-specific helper when the input has at least\n  `Count` elements; otherwise run a dedicated small-input helper, then return. Do not put the length\n  check in the outer guard and fall through to repeat dispatch at narrower widths. Keeping each\n  supported-width block self-contained lets the JIT remove unsupported blocks and avoids redundant\n  work on common small inputs.\n- **Loads and stores:** prefer span-based `Vector128.Create(span)` and `CopyTo`; the JIT keeps them\n  efficient and they require no pinning or reference arithmetic. Unsafe loads and stores are largely\n  unnecessary. When a path genuinely must walk a buffer by managed reference, use the element-offset\n  `LoadUnsafe(ref T, nuint)` and `StoreUnsafe` overloads rather than pointers or manually advanced\n  references.\n- **Empty inputs:** in a reference-based path, obtain the starting reference with\n  `MemoryMarshal.GetReference(span)` or `MemoryMarshal.GetArrayDataReference(array)`, not by indexing\n  element `0`.\n- **Unsupported element types:** the fixed-width vectors support primitive numeric element types,\n  not `char` or `bool`. Reinterpret with `MemoryMarshal.Cast` or `As\u003CTFrom, TTo>`; reinterpretation\n  changes only the type, not the bits. Keep Boolean data as `0` or `1` and characters as valid\n  UTF-16, normalizing results before storing when necessary.\n- **Offsets:** prove the input contains a full vector before subtracting `Count` or converting an\n  index to `nuint`; otherwise a negative value becomes a huge unsigned offset.\n- **Managed references:** do not form references before the start or past the end of a span,\n  including a one-past-end reference. The runtime permits a non-dereferenced managed pointer exactly\n  one past an object or array, but this guidance intentionally prohibits the pattern because it is\n  fragile and easy to misuse. Keep the base reference in range and express traversal with an element\n  offset.\n- **Remainders:** cover every length, including `0`, `Count - 1`, `Count`, `Count + 1`, and\n  nonmultiples of each width. Once the input contains a full vector, keep the tail vectorized by\n  reprocessing the last full vector. An idempotent operation can fold that overlap in directly. A\n  non-idempotent operation must use `ConditionalSelect` to replace repeated lanes with the\n  operation's identity before folding them in. This is the JIT-recognized general pattern; it can\n  reduce a zero-identity selection to a bitwise mask while retaining broader optimization\n  opportunities. For in-place transforms, preserve the original tail values before overlapping\n  stores and write only valid results.\n- **Buffer overlap:** choose a traversal direction or staging strategy that prevents stores from\n  corrupting values not yet loaded.\n- **Numeric behavior:** account for floating-point reassociation, NaN and signed-zero semantics,\n  checked or unchecked integer overflow, and endianness where the algorithm depends on byte order.\n  `Native` and `Estimate` operations can intentionally relax precision or IEEE edge-case behavior;\n  use them only when the contract permits it and measurements justify them.\n\nThe official guidance contains the complete dispatch, small-input, unrolling, and remainder\ntemplates; use those for the full implementation. The following excerpt illustrates only the inner\nsafe `Vector128\u003CT>` loop for an in-place elementwise transform, after its self-contained dispatch\nblock has established at least one full vector. `Transform` represents the operation being\nimplemented:\n\n```csharp\nSpan\u003Cint> tail = data.Slice(data.Length - Vector128\u003Cint>.Count);\nVector128\u003Cint> end = Vector128.Create\u003Cint>(tail);\nSpan\u003Cint> remaining = data;\n\nwhile (remaining.Length >= Vector128\u003Cint>.Count)\n{\n    Vector128\u003Cint> values = Vector128.Create\u003Cint>(remaining);\n    Transform(values).CopyTo(remaining);\n    remaining = remaining.Slice(Vector128\u003Cint>.Count);\n}\n\nif (!remaining.IsEmpty)\n{\n    Transform(end).CopyTo(tail);\n}\n```\n\nThe early `end` load preserves original values before overlapping stores. For a read-only reduction,\nload the same final span after the main loop and use `ConditionalSelect` to replace already-processed\nlanes with the operation's identity. Do not substitute `LoadUnsafe`\u002F`StoreUnsafe` or a scalar\nepilogue merely to avoid span bounds checks.\n\n## Testing checklist\n\n- Compare the optimized implementation with the scalar contract across boundary lengths,\n  randomized values, empty inputs, supported overlap, and numeric edge cases. Cover every\n  implemented width and the scalar path with inputs both large enough and too small to benefit.\n- Exercise every implemented width and the scalar fallback in separate processes. On x86\u002Fx64\n  CoreCLR, `DOTNET_EnableAVX2=0` disables AVX2 and `DOTNET_EnableHWIntrinsic=0` disables hardware\n  intrinsics. Use the repository's normal test command and do not change these process-wide\n  settings inside a unit test. These settings do not change code already compiled as ReadyToRun or\n  ahead of time, so confirm the target code is JIT-compiled when using them to force a path.\n- For unsafe loads and stores, use guard-page or equivalent boundary tests when available. Put the\n  inaccessible page after the buffer for forward iteration and before it for backwards iteration,\n  and include nonmultiple lengths. An ordinary array allocation does not reliably expose an\n  out-of-bounds read.\n\n## Benchmarking\n\nUse BenchmarkDotNet to measure representative small and large inputs before keeping the added\ncomplexity. Compare scalar, `Vector128\u003CT>`, and each wider implemented path in the same run. Small\ninputs can be slower because setup dominates, and speedups are rarely the theoretical vector-width\nmultiple because memory throughput, alignment, and latency still apply. Report throughput or time\nwith noise context and, when relevant, generated code size or instruction counts. Control allocation\nalignment for stable measurements or randomize it to observe the distribution. A wider vector is\nnot automatically faster.\n\nIf the project cannot target the required framework, run the relevant architecture, or execute the\nfallback configuration, state exactly which path remains unverified. Do not claim success from a\ndefault-hardware test alone.\n\n## Completion contract\n\n- **Authoring:** leave the scalar contract covered by tests; identify the framework or SIMD layer\n  selected; report measurements for the representative workload; name any architecture or fallback\n  path that could not be exercised.\n- **Review:** report only concrete findings, ordered by correctness, memory safety, portability,\n  tests, then performance evidence. If none remain, say so directly.\n- Do not call an optimization complete when it only builds, only passes on the current machine, or\n  has no comparison against the scalar baseline.\n\n## Review checklist\n\nReview in this order:\n\n1. Scalar-contract equivalence, including signed zero, NaN, overflow, and relevant endianness\n2. Reuse of an existing accelerated framework API\n3. Tail correctness for idempotent versus non-idempotent work\n4. Memory safety, unsigned offset arithmetic, empty inputs, and overlapping buffers\n5. Portable dispatch and behaviorally equivalent fallbacks\n6. Tests that force each width and the scalar path\n7. Benchmarks that justify explicit SIMD and additional widths\n",{"name":5,"description":7},"master","b30b1c1908dbc56a8184128f18803053d762ccb5","opencode\u002Fdeniz-dotnet-general\u002Fskills\u002Fdeniz-dotnet-general.vectorization",0,"https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins","https:\u002F\u002Fgithub.com\u002FBlind-Striker","denizirgin.com","VP of Engineering - \r\n@armutcom @homeruntech @prontopro\r\n |  Speaker \r\n@CodeFiction\r\n, http:\u002F\u002Fcodefiction.tech | Developer, gamer, blogger, maiden iron fan","2026-11-13 00:23:18",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add Blind-Striker\u002Fagent-skills-and-plugins --skill deniz-dotnet-general.vectorization",{"tool":39,"command":40},"bunx","bunx skills add Blind-Striker\u002Fagent-skills-and-plugins --skill deniz-dotnet-general.vectorization",{"tool":42,"command":43},"pnpm","pnpm dlx skills add Blind-Striker\u002Fagent-skills-and-plugins --skill deniz-dotnet-general.vectorization","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fopencode\u002Fdeniz-dotnet-general\u002Fskills\u002Fdeniz-dotnet-general.vectorization\u002FSKILL.md",true,[],[48,55,62,69,76,83],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M79YGWBJ52PHMSTKJ3RJ3AWR","writing-tunit-tests","Write tests in TUnit — the attributes, the async assertion model, data-driven cases, lifecycle hooks, and the parallelism rules. USE FOR writing or fixing a test in a TUnit project, translating a test idea into TUnit syntax, or reviewing TUnit test code for API mistakes. DO NOT USE FOR deciding what to test, judging whether a test is any good, or running tests.","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fskills\u002Fdeniz-dotnet-general\u002Fwriting-tunit-tests\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M79YGR3Y9RCRS6TD71D0KKP8","dotnet-walls","Walls for a .NET change in a repository without its own quality gates — pinned, ephemeral analyzer, format, clone and scanner commands for the verify step, their identity ratchet, and the build traps. Use during a walls preflight, or when a worker's verify step needs .NET quality gates.","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fskills\u002Fdeniz-dotnet-general\u002Fdotnet-walls\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M79YGFNJBST7D803NQW3SBYM","csharp-15","Write, review and modernize C# 15 (.NET 11) code correctly. Covers every C# 15 feature - collection expression arguments (`[with(...), ...]`), union types (`union`, `[Union]`, `IUnion`, custom unions), closed class hierarchies (`closed`), extension indexers, labeled `break`\u002F`continue`, non-virtual static interface members on runtimes without default interface methods, and the preview memory-safety model (`unsafe(expr)`, `safe`, requires-unsafe members). Use when code targets net11.0 or LangVersion 15\u002Fpreview, when the user mentions any of these features, when replacing marker interfaces, OneOf-style results, `goto`\u002Fflag loop exits or capacity\u002Fcomparer boilerplate, or when the build reports CS8121, CS8509, CS8655, CS8510, CS0121\u002FCS0411 on a `with(...)` element, CS9370, CS9380 or CS9363.","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fskills\u002Fdeniz-dotnet-general\u002Fcsharp-15\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M79YGAV031473HSX6FN87ENH","verified-waves","Run one task as verified multi-agent waves. Plans the fan-out, sends findings to fresh skeptics, keeps a ledger, and reports N of M done; small tasks stay solo.","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fskills\u002Fdeniz-coordination\u002Fverified-waves\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M79YG6QXNND4WRPP3GATSY8J","harness","The bundled dispatch harness, which runs a task file on worker models through OpenCode or Codex, fences their writes, escalates their asks to the operator, settles each result from git and its verify commands, and records every step in a ledger. Its commands, task file, modes and write fence, configuration, runs and their recovery, ledger and exit codes.","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fskills\u002Fdeniz-coordination\u002Fharness\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M79YG2P5S91SDTA5912XJ4FV","delegating-to-worker-models","Use when you hand off implementation, a plan, research a decision will rely on, or a review, and you brief the worker and check what comes back. A subagent that searches, reads in parallel, or keeps your context small is ordinary tool use. Covers what stays with you, the mechanism, assigning work by strength, the walls preflight, prompts without contradictions, the worker contract and model profiles, evidence from your own run, review, and the fix-up loop.","https:\u002F\u002Fgithub.com\u002FBlind-Striker\u002Fagent-skills-and-plugins\u002Fblob\u002Fmaster\u002Fskills\u002Fdeniz-coordination\u002Fdelegating-to-worker-models\u002FSKILL.md",[],[]]