PyTorch PR Review Skill

Usage Modes

No Argument

/pr-review

What would you like me to review?

• A PR number or URL (e.g.,

  /pr-review 12345

  )
• A local branch (e.g.,

  /pr-review branch

  )

Local CLI Mode

/pr-review 12345
/pr-review https://github.com/pytorch/pytorch/pull/12345

/pr-review 12345 detailed

gh

# Get PR details
gh pr view <PR_NUMBER> --json title,body,author,baseRefName,headRefName,files,additions,deletions,commits

# Get the diff
gh pr diff <PR_NUMBER>

# Get PR comments
gh pr view <PR_NUMBER> --json comments,reviews

Local Branch Mode

main

/pr-review branch
/pr-review branch detailed

# Get current branch name
git branch --show-current

# Get list of changed files compared to main
git diff --name-only main...HEAD

# Get full diff compared to main
git diff main...HEAD

# Get commit log for the branch
git log main..HEAD --oneline

# Get diff stats (files changed, insertions, deletions)
git diff --stat main...HEAD

• The "Summary" should describe what the branch changes accomplish based on commit messages and diff
• Use the current branch name in the review header instead of a PR number
• All other review criteria apply the same as PR reviews

GitHub Actions Mode

Review Workflow

Step 1: Fetch PR Information

gh

1. PR metadata (title, description, author)
2. List of changed files
3. Full diff of changes
4. Existing comments/reviews
5. Fetch associated issue information when applicable

Step 2: Analyze Changes

1. Identify the purpose of the change from title/description/issue
2. Group changes by type (new code, tests, config, docs)
3. Note the scope of changes (files affected, lines changed)

Step 3: Deep Review

• Code quality and design
• Testing adequacy
• Security considerations
• Thread safety and concurrency (Python, C++, CPython C API, NoGIL)
• Performance implications
• Any behavior change not expected by author

Step 4: Check Backward Compatibility

• What constitutes a BC-breaking change
• Required deprecation patterns
• Common BC pitfalls

Step 5: Formulate Review

Review Areas

Output Format

## PR Review: #<number>
<!-- Or for local branch reviews: -->
## Branch Review: <branch-name> (vs main)

### Summary
Brief overall assessment of the changes (1-2 sentences).

### Code Quality
[Issues and suggestions, or "No concerns" if none]

### API Design
[Flag new patterns, internal-access flags, or broader implications if any. Otherwise omit this section.]

### Testing
- [ ] Tests exist for new functionality
- [ ] Edge cases covered
- [ ] Tests follow PyTorch patterns (TestCase, assertEqual)
[Additional testing feedback]

### Security
[Issues if any, or "No security concerns identified"]

### Thread Safety
[Threading concerns if any, or "No thread safety concerns"]

### Backward Compatibility
[BC concerns if any, or "No BC-breaking changes"]

### Performance
[Performance concerns if any, or "No performance concerns"]

### Recommendation
**Approve** / **Request Changes** / **Needs Discussion**

[Brief justification for recommendation]

Specific Comments (Detailed Review Only)

### Specific Comments
- `src/module.py:42` - Consider extracting this logic into a named function for clarity
- `test/test_feature.py:100-105` - Missing test for error case when input is None
- `torch/nn/modules/linear.py:78` - This allocation could be moved outside the loop

Key Principles

1. No repetition - Each observation appears in exactly one section. Never repeat the same issue, concern, or suggestion across multiple sections. If an issue spans categories (e.g., a security issue that also affects performance), place it in the most relevant section only.
2. Focus on what CI cannot check - Don't comment on formatting, linting, or type errors
3. Be specific - Reference file paths and line numbers
4. Be actionable - Provide concrete suggestions, not vague concerns
5. Be proportionate - Minor issues shouldn't block, but note them
6. Assume competence - The author knows PyTorch; explain only non-obvious context

Files to Reference

• CLAUDE.md

  - Coding style philosophy and testing patterns

• CONTRIBUTING.md

  - PR requirements and review process

• torch/testing/_internal/common_utils.py

  - Test patterns and utilities

• torch/testing/_internal/opinfo/core.py

  - OpInfo test framework