Does it provide templates for examples?

Yes, it formats examples using the `Examples::` directive and the `>>>` Python prompt to ensure they are both readable and testable.

What is the standard indentation for parameter descriptions?

It follows PyTorch's convention of using a 2-space indentation for continuation lines within the 'Args' and 'Keyword Args' sections.

How does it manage cross-references between modules?

The skill uses Sphinx roles like `:class:`, `:func:`, and `:meth:` with the tilde prefix to create clean, clickable links in the generated documentation.

Does this skill support LaTeX for math formulas?

Yes, it uses raw strings and Sphinx math directives to ensure mathematical expressions and tensor shapes render correctly in PyTorch documentation.

Can it handle C++ bound functions in PyTorch?

Yes, it includes specific patterns for using `_add_docstr` and `add_docstr_all` to document C-bound functions and aliases.

PyTorch Docstring Formatter

Name: PyTorch Docstring Formatter
Author: pytorch

bypytorch

•

96,867

数据科学与机器学习

Generates and formats standardized PyTorch docstrings following Sphinx and reStructuredText conventions.

关于

The PyTorch Docstring skill provides Claude with domain-specific guidance for writing high-quality documentation within the PyTorch ecosystem. It automates the complex formatting required for PyTorch functions, including raw string usage for LaTeX math, specific signature structures, and standardized sections for arguments, keyword arguments, and examples. By following the conventions found in torch/_tensor_docs.py, this skill ensures that all documentation is consistent, cross-referenced correctly, and ready for Sphinx rendering, making it an essential tool for PyTorch contributors and library maintainers.

主要功能

Automates function signature formatting with positional and keyword-only arguments
Creates standardized Args, Keyword Args, and Returns sections with proper indentation
Generates LaTeX-compliant mathematical formulas using Sphinx math directives
Supports specialized docstring patterns for C-bound functions and in-place variants
96,867 GitHub stars
Implements intelligent cross-referencing for PyTorch classes, functions, and methods

使用场景

Documenting new PyTorch native or C++ bound functions with proper reST syntax
Generating complex tensor shape documentation and mathematical explanations for ML operations
Updating existing codebase documentation to meet PyTorch's strict formatting standards

关于

主要功能

Automates function signature formatting with positional and keyword-only arguments
Creates standardized Args, Keyword Args, and Returns sections with proper indentation
Generates LaTeX-compliant mathematical formulas using Sphinx math directives
Supports specialized docstring patterns for C-bound functions and in-place variants
96,867 GitHub stars
Implements intelligent cross-referencing for PyTorch classes, functions, and methods

使用场景

Documenting new PyTorch native or C++ bound functions with proper reST syntax
Generating complex tensor shape documentation and mathematical explanations for ML operations
Updating existing codebase documentation to meet PyTorch's strict formatting standards