review-pr

Analyze GitHub pull requests with diffs, comments, issues, and local code context.

5.7k|578|Updated Jan 26, 2022
One-click install
npx skills add https://github.com/dlt-hub/dlt --skill review-pr-dlt-hub
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-pr
Source: https://github.com/dlt-hub/dlt/tree/main/.claude/skills/review-pr
Command: npx skills add https://github.com/dlt-hub/dlt --skill review-pr-dlt-hub

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

PR reviews are time-consuming and error-prone when context is scattered. This Skill analyzes diffs, comments, issues, and local code context to deliver a comprehensive, reproducible PR review.

Core Features & Use Cases

  • Structured PR metadata extraction: title, author, state, base and head refs, URL, and labels
  • Diffs and discussions aggregation: pull request diffs, comments, and related issues
  • Context-aware recommendations: actionable review items with rationale and risk assessment
  • Use Case: Streamline code review cycles for large PRs by summarizing key changes and potential conflicts

Quick Start

Provide a PR URL or number to start analysis and generate a structured review.

Frequently Asked Questions about review-pr

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze a GitHub pull request with local code context and related issues?

To analyze a GitHub pull request, provide the PR URL or number. The Skill extracts diffs, comments, and related issues, then cross-references local code context to deliver a comprehensive review with risk assessments.

What is the best way to automate PR reviews for large diffs and scattered discussions?

Automating PR reviews is best handled by aggregating pull request metadata, diffs, and related issues into a structured summary. This approach identifies potential conflicts and highlights requested changes with clear rationale.

Does automated PR review extract metadata like base and head refs along with the diff?

Automated PR review does extract structured metadata including title, author, state, base and head refs, and labels. It aggregates this metadata with diffs and discussions to provide context-aware recommendations.

Can I generate a code review verdict that highlights risks and suggested improvements?

You can generate a concise code review verdict that highlights risks, changes requested, and suggested improvements. The verdict is produced by analyzing the diff and related discussions with justification.

How do I streamline code review cycles for pull requests across multiple repositories?

Streamlining code review cycles across repositories requires applying automated analysis to each PR. The Skill processes reading metadata, diffs, and associated discussions to produce reproducible, comprehensive reviews.

What context is needed to perform context-aware recommendations on a pull request?

Context-aware recommendations require the pull request diff, comments, related issues, and local code context. Aggregating these elements allows the review to provide actionable items with rationale and risk assessment.