--- title: "pytorch skills · skilld" canonical_url: "https://skilld.dev/gh/pytorch" meta: description: "19 agent skills published by pytorch on skilld. pytorch, Documentation, Debugging." "og:description": "19 agent skills published by pytorch." "og:title": "pytorch on skilld" --- ` ![Avatar for pytorch](https://skilld.dev/_img/avatar?url=https%3A%2F%2Fgithub.com%2Fpytorch.png) # **pytorch** [@pytorch](https://github.com/pytorch)org 19 skills 104k where the eigens are valued Mostly·pytorch, Documentation, Debugging [GitHub](https://github.com/pytorch) [Website](https://pytorch.org) ## Skills ### [pytorch/pytorch](https://skilld.dev/gh/pytorch/pytorch) Tensors and Dynamic neural networks in Python with strong GPU acceleration 19 skills 104k `npx skilld add pytorch/pytorch` - [ **/add-uint-support**104k Add unsigned integer (uint) type support to PyTorch operators by updating AT\_DISPATCH macros. Use when adding support for uint16, uint32, uint64 types to operators, kernels, or when user mentions enabling unsigned types, barebones unsigned types, or uint support. /add-uint-support by pytorch](https://skilld.dev/gh/pytorch/pytorch/add-uint-support) - [ **/aoti-debug**104k Debug AOTInductor (AOTI) errors and crashes. Use when encountering AOTI segfaults, device mismatch errors, constant loading failures, or runtime errors from aot\_compile, aot\_load, aoti\_compile\_and\_package, or aoti\_load\_package. /aoti-debug by pytorch](https://skilld.dev/gh/pytorch/pytorch/aoti-debug) - [ **/at-dispatch-v2**104k Convert PyTorch AT\_DISPATCH macros to AT\_DISPATCH\_V2 format in ATen C++ code. Use when porting AT\_DISPATCH\_ALL\_TYPES\_AND\*, AT\_DISPATCH\_FLOATING\_TYPES\*, or other dispatch macros to the new v2 API. For ATen kernel files, CUDA kernels, and native operator implementations. /at-dispatch-v2 by pytorch](https://skilld.dev/gh/pytorch/pytorch/at-dispatch-v2) - [ **/ci-metrics**104k Query PyTorch CI, GitHub Actions, HUD, Grafana, and infrastructure metrics. Use when users ask about CI duration, job failures, queue times, workflow trends, runner health, dashboard data, or PyTorch infrastructure metrics. /ci-metrics by pytorch](https://skilld.dev/gh/pytorch/pytorch/ci-metrics) - [ **/cuda-index-width**104k Choose 32-bit vs 64-bit index math in PyTorch CUDA kernels. Use when fixing large-tensor indexing overflows, deciding whether to use int64\_t, canUse32BitIndexMath, CUDA\_KERNEL\_LOOP\_TYPE, or AT\_DISPATCH\_INDEX\_TYPES, and when considering binary-size or performance impact of index-type templating. /cuda-index-width by pytorch](https://skilld.dev/gh/pytorch/pytorch/cuda-index-width) - [ **/distributed-triage**104k Sub-triages issues in the oncall:distributed queue by assigning distributed module labels, routing to sub-oncalls, and marking triaged. Use when an issue has been routed to oncall:distributed and needs second-level triage. /distributed-triage by pytorch](https://skilld.dev/gh/pytorch/pytorch/distributed-triage) - [ **/docstring**104k Write docstrings for PyTorch functions and methods following PyTorch conventions. Use when writing or updating docstrings in PyTorch code. /docstring by pytorch](https://skilld.dev/gh/pytorch/pytorch/docstring) - [ **/document-public-apis**104k Document undocumented public APIs in PyTorch by removing functions from coverage\_ignore\_functions and coverage\_ignore\_classes in docs/source/conf.py, running Sphinx coverage, and adding the appropriate autodoc directives to the correct .md or .rst doc files. Use when a user asks to remove functions from conf.py ignore lists. /document-public-apis by pytorch](https://skilld.dev/gh/pytorch/pytorch/document-public-apis) - [ **/fix-issue**104k Fix bugs reported in PyTorch GitHub issues by reproducing, root-causing, and implementing a fix in the local working tree. Use when the user asks to fix a PyTorch GitHub issue. /fix-issue by pytorch](https://skilld.dev/gh/pytorch/pytorch/fix-issue) - [ **/ghstack-ci**104k Manage CI for PyTorch ghstack stacks by running CI where its results are useful now and deferring other PRs with \[no-ci\]. Use when creating, submitting, updating, restacking, or landing a stack of PRs. /ghstack-ci by pytorch](https://skilld.dev/gh/pytorch/pytorch/ghstack-ci) - [ **/kineto-release**104k Update the third\_party/kineto submodule in PyTorch to the latest commit from this kineto repo and commit the change. Use when updating the kineto submodule hash for a release. /kineto-release by pytorch](https://skilld.dev/gh/pytorch/pytorch/kineto-release) - [ **/metal-kernel**104k Write Metal/MPS kernels for PyTorch operators. Use when adding MPS device support to operators, implementing Metal shaders, or porting CUDA kernels to Apple Silicon. Covers native\_functions.yaml dispatch, host-side operators, and Metal kernel implementation. /metal-kernel by pytorch](https://skilld.dev/gh/pytorch/pytorch/metal-kernel) - [ **/pr-review**104k Review PyTorch pull requests for code quality, test coverage, security, and backward compatibility. Use when reviewing PRs, when asked to review code changes, or when the user mentions "review PR", "code review", or "check this PR". /pr-review by pytorch](https://skilld.dev/gh/pytorch/pytorch/pr-review) - [ **/pr-review-readiness**104k Non-interactive wrapper over pr-review that emits JSON assessing whether a PR is ready for a human maintainer's time. /pr-review-readiness by pytorch](https://skilld.dev/gh/pytorch/pytorch/pr-review-readiness) - [ **/pt2-bug-basher**104k Debug PyTorch 2 compiler stack failures including Dynamo graph breaks, Inductor codegen errors, AOTAutograd crashes, and accuracy mismatches. Use when encountering torch.compile errors, BackendCompilerFailed exceptions, recompilation issues, Triton kernel failures, FX graph problems, or when the user mentions debugging PT2, Dynamo, Inductor, or compiled model issues. /pt2-bug-basher by pytorch](https://skilld.dev/gh/pytorch/pytorch/pt2-bug-basher) - [ **/pyrefly-type-coverage**104k Migrate a file to use stricter Pyrefly type checking with annotations required for all functions, classes, and attributes. /pyrefly-type-coverage by pytorch](https://skilld.dev/gh/pytorch/pytorch/pyrefly-type-coverage) - [ **/scrub-issue**104k Fetch, analyze, reproduce, and minimize GitHub issue reproductions. Use only when the user asks to check whether a GitHub issue reproduces, minimize a GitHub issue's repro, analyze a GitHub bug report, or scrub/triage a GitHub issue for reproducibility. Do not use for standalone reproduction requests without a GitHub issue. /scrub-issue by pytorch](https://skilld.dev/gh/pytorch/pytorch/scrub-issue) - [ **/skill-writer**104k Guide users through creating Agent Skills for Claude Code. Use when the user wants to create, write, author, or design a new Skill, or needs help with SKILL.md files, frontmatter, or skill structure. /skill-writer by pytorch](https://skilld.dev/gh/pytorch/pytorch/skill-writer) - [ **/triaging-issues**104k Triages GitHub issues by routing to oncall teams, applying labels, and closing questions. Use when processing new PyTorch issues or when asked to triage an issue. /triaging-issues by pytorch](https://skilld.dev/gh/pytorch/pytorch/triaging-issues) Skills published by pytorch. Checked GitHub just now