code-review-parallel

Coordinate parallel pull request reviews across GitHub, GitLab, and Azure DevOps.

5|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/kenlck/skills --skill code-review-parallel
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-review-parallel
Source: https://github.com/kenlck/skills/tree/main/ken-swe/skills/code-review-parallel
Command: npx skills add https://github.com/kenlck/skills --skill code-review-parallel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Batch-review multiple PRs across GitHub, GitLab, and Azure DevOps by distributing work to dedicated agents while maintaining a consolidated summary.

Core Features & Use Cases

  • Thin orchestrator that fans out the single-PR code-review pipeline across many PRs/MRs.
  • One agent per PR to keep reviews scoped and parallelizable.
  • Supports GitHub, GitLab, and Azure DevOps, read-only by design, with per-repo CLAUDE.md conventions.

Quick Start

Start a parallel review by supplying a list of PR URLs or IDs to spawn per-PR agents and generate a consolidated report.

Frequently Asked Questions about code-review-parallel

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

FAQPage Schema
How do I review multiple pull requests across different repositories at once?▼

You can batch-review multiple pull requests by distributing work to dedicated agents. This orchestrator spawns one agent per PR to keep reviews scoped and parallelizable, then generates a consolidated summary report across all repositories.

Can I run parallel code reviews across GitHub, GitLab, and Azure DevOps simultaneously?▼

Yes, parallel code reviews support GitHub, GitLab, and Azure DevOps simultaneously. The pipeline operates read-only by design and applies per-repo CLAUDE.md conventions to maintain consistent analysis across all three platforms.

How do I start a batch code review for a release sweep?▼

To start a batch code review, supply a list of PR URLs or IDs to the orchestrator. It fans out the single-PR review pipeline across all provided items, spawning per-PR agents to execute a round-robin review workflow.

Do I need a CLAUDE.md file in each repository to perform multi-repo audits?▼

Yes, per-repo CLAUDE.md files are required to perform multi-repo audits. These files provide the repository-specific conventions that guide the read-only analysis and ensure each per-PR agent respects the project's standards.

What is the best way to handle diff filtering during a batch pull request review?▼

Diff filtering is handled within the round-robin review pipeline. The thin orchestrator delegates diff filtering and read-only analysis to each dedicated per-PR agent, ensuring only relevant changes are processed in the consolidated report.

Are there limitations to using parallel agents for code review workflows?▼

The primary limitation is that parallel code review workflows are read-only by design, meaning agents cannot modify or approve pull requests. The orchestrator strictly coordinates analysis and generates summaries without executing write operations.