Torch Xla | Agent Skills Atlas

About this skill

Expert guide for PyTorch/XLA (torchxla) development targeting TPU and XLA-compatible devices. Use this skill whenever the user asks about: torchxla, PyTorch on TPU, XLA training, SPMD parallelism, distributed TPU training, XLA checkpointing, MpDeviceLoader, xm.markstep, torchxla.launch, FSDPv2 on XLA, gradient sharding, XLA compilation, lazy tensor execution, or migrating GPU PyTorch code to TPU. Trigger even for partial mentions like "TPU training", "XLA device", "xla backend", "run on TPU", or "SPMD mesh". This skill covers torchxla 2.9.0 (latest stable, requires torch==2.9.0, Python 3.10–3.13).

Quick start

Copy the install prompt. Use the green button above this README.
Send it to your agent. Paste it into your coding agent’s chat and send.
Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.

Installation guide →

Review, license & listing details

Review the source before using this skill

This is an imported listing. We haven’t certified its safety, license, or compatibility. Skills may include executable scripts.

Automated security scan: Partial

Only SKILL.md was scanned. Supporting files and scripts were not checked. No safety grade has been assigned.

0 potential findings · Scanned 11/4/2026 · Scanner 0.2.1

Content SHA-256: 7e2a69afcd3c9c28ea404f7b425f323efd877b14adc5374fd897741b1cafb2be

About scan coverage and limitations

Original listing metadata

Original name

torch-xla

Original description

Expert guide for PyTorch/XLA (torch_xla) development targeting TPU and XLA-compatible devices. Use this skill whenever the user asks about: torch_xla, PyTorch on TPU, XLA training, SPMD parallelism, distributed TPU training, XLA checkpointing, MpDeviceLoader, xm.mark_step, torch_xla.launch, FSDPv2 on XLA, gradient sharding, XLA compilation, lazy tensor execution, or migrating GPU PyTorch code to TPU. Trigger even for partial mentions like "TPU training", "XLA device", "xla backend", "run on TPU", or "SPMD mesh". This skill covers torch_xla 2.9.0 (latest stable, requires torch==2.9.0, Python 3.10–3.13).

About this listing

Category: Engineering
Reported license: Not specified
Listing language (estimated): English
Review status: Not individually reviewed
Agent compatibility: Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.