Skip to content
TrustedSkillsbeta
Back to all skills

Aoti Debug

Official
pytorch·vlatest·testing

Automatically identifies and suggests fixes for common PyTorch training loop errors using AI debugging techniques.

Repository

Install on your platform

Showing Claude Code. One command installs this skill into any of the agents below.

  1. 1

    Run this in your project

    terminal
    $ npx skills add pytorch/pytorch --skill aoti-debug -a claude-code

  2. Uses the open-source skills CLI and needs Node.js. Add -g to install it for your user instead of this project.

What we know about this skill

Publisher

pytorch

Matched to the vendor's own GitHub organisation.

Install target

Live repository

Not pinned. Installing fetches whatever the repository holds at the time you run the command.

No human has reviewed this code. The checks below are a static scan of the files at one commit — nobody has run this skill or judged whether it works. A skill runs with whatever access you give your agent, so read the source before you install it.

Automated safety checks

scanned

This skill passed all 6 checks. We read its files at one commit and looked for the things that make a skill dangerous to run — not for whether it works.

  • Declares what it does: passed

    SKILL.md parses and declares a name and a description.

    SKILL.md exists, its frontmatter parses, and it declares a name and a description.

  • No undeclared network calls: passed

    Makes no network calls.

    Every host the skill dials — from its scripts, its config endpoints or the commands it tells the agent to run — is either well-known package and API infrastructure or declared in its own frontmatter. Links in documentation are not counted.

  • No obfuscated payloads: passed

    No encoded or obfuscated payloads.

    No base64, hex or compressed blob is decoded and then executed, and no encoded shell commands.

  • No credential access: passed

    Reads no keys, tokens, keychains or credential files.

    Neither the code nor the SKILL.md instructions read SSH keys, cloud credentials, keychains, token stores, browser cookie databases or dotfiles that hold secrets.

  • No pipe-to-shell installers: passed

    No pipe-to-shell or download-and-run installers.

    No `curl … | sh` pattern, in the code or in the instructions, that runs code fetched at run time.

  • Pinned to a commit: passed

    Scanned pytorch/pytorch at edd4fc2d34.

    The exact upstream commit SHA that was scanned is recorded, so the result refers to specific bytes.

Commit
edd4fc2d34
Manifest
.claude/skills/aoti-debug/SKILL.md
Scanned
2 files, 391 lines
This is a static scan, not a review. It reads the skill's files; it does not run them, and it cannot tell you whether the skill is any good or whether the code does what its description says. A skill runs with whatever access you give your agent.
What each check looks for →

About this skill

What it does

The aoti-debug skill enables developers to debug models compiled with AOTInductor, providing visibility into the compilation process and runtime behavior. It facilitates troubleshooting by allowing inspection of generated code and execution traces within the PyTorch ecosystem.

When to use it

  • You are encountering runtime errors in models optimized via AOTInductor and need to isolate the failure point.
  • You want to inspect the intermediate representation or lowered code produced during the AOT compilation pipeline.
  • Your team is debugging performance bottlenecks specific to compiled PyTorch graphs rather than eager mode execution.
  • You need to validate that custom operators or kernels are correctly integrated into the AOTInductor build.

Key capabilities

  • Debugging support for models compiled with AOTInductor.
  • Inspection of compilation artifacts and generated code.
  • Runtime tracing for optimized PyTorch workflows.

Example prompts

  • "Run a debug session on this model compiled with AOTInductor to identify the source of the CUDA error."
  • "Generate a trace of the execution flow for an AOT-compiled graph to check operator fusion logic."
  • "Inspect the lowered code produced by AOTInductor for this custom layer to verify kernel registration."

Tips & gotchas

Ensure your PyTorch installation includes the latest torch.compile and AOTInductor features, as debugging tools depend on these being active. This skill is specifically designed for compiled workflows; it may not provide useful output when running models in standard eager mode without explicit compilation flags.