cmd-pr-sweep

Review changes for test gaps, simplification, naming consistency, reuse opportunities, and TODO quality

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "cmd-pr-sweep" with this command: npx skills add olshansk/agent-skills/olshansk-agent-skills-cmd-pr-sweep

Code Review Sweep

Review all recent changes (staged, unstaged, or a specified branch diff) across five dimensions. Focus on actionable findings — skip nitpicks.

Instructions

  1. Identify the change scope — ask the user if unclear:

    • Uncommitted changes: git diff HEAD
    • Branch diff: git diff main...HEAD
    • Specific files: user-provided list
  2. Read all changed files in full before reviewing. Understand the existing patterns, not just the diff.

  3. Review each dimension below. For each finding, cite the file and line.

Review Dimensions

1. Test Gaps

  • Are there new code paths without corresponding tests?
  • Are edge cases covered (empty inputs, error paths, boundary values)?
  • Do existing tests still pass with the changes?
  • Are there testable utilities or helpers that lack unit tests?

2. Simplification

  • Can any conditional branches be collapsed or removed?
  • Are there unnecessary abstractions (wrappers, factories, indirection) for single-use cases?
  • Can complex expressions be broken into named variables for clarity?
  • Are there redundant checks or defensive code that can't trigger?

3. Naming Consistency

  • Do new names follow existing conventions in the codebase?
  • Are abbreviations used consistently (don't mix repo/repository, config/configuration)?
  • Do function names accurately describe what they do?
  • Are similar concepts named similarly across files?

4. Reuse Opportunities

  • Is there duplicated logic that could use an existing helper?
  • Are there patterns repeated 3+ times that should be extracted?
  • Can existing utilities, constants, or types be reused instead of redefined?
  • Are there opportunities to consolidate similar functions?

5. TODO & Comment Quality

  • Do TODOs follow the project's prefix convention (TODO, TODO_IMPROVE, TODO_TECHDEBT, etc.)?
  • Does each TODO include what and why?
  • Are there stale or resolved TODOs that should be removed?
  • Are comments explaining "why" rather than "what"?
  • Are there missing TODOs for known shortcuts or deferred work?

Output Format

For each dimension, output:

### [Dimension Name]

- **[file:line]** — Finding description
  Suggestion: ...

- **[file:line]** — Finding description
  Suggestion: ...

If a dimension has no findings, output: No issues found.

Final Summary

End with a prioritized list of the top 3-5 most impactful changes, ranked by effort-to-value ratio.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

session-commit

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

skills-dashboard

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

cmd-rss-feed-generator

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

cmd-idiot-proof-docs

No summary provided by upstream source.

Repository SourceNeeds Review