address-pr-review

Orchestrate parallel PR review workflows with finding triage, automated fixes, and CI monitoring.

Updated Apr 29, 2024
One-click install
npx skills add https://github.com/takaokanbe/dotfiles --skill address-pr-review-takaokanbe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: address-pr-review
Source: https://github.com/takaokanbe/dotfiles/tree/main/config/claude/skills/address-pr-review
Command: npx skills add https://github.com/takaokanbe/dotfiles --skill address-pr-review-takaokanbe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually managing PR review cycles—collecting feedback, triaging findings, fixing issues, and monitoring CI—is repetitive, error-prone, and easy to get stuck in infinite loops without proper guardrails.

Core Features & Use Cases

  • Parallel Review Orchestration: Runs multiple review agents (codex review, adversarial review, pr-review-toolkit) in parallel to collect comprehensive feedback in a single pass.
  • Intelligent Finding Triage: Cross-validates findings across review sources, applies session-aware severity adjustment, and filters out false positives to prioritize critical issues.
  • Iterative Fix & CI Monitoring: Automatically fixes triaged issues, runs local validation, monitors CI status, and retries failed CI jobs up to a configurable limit before escalating to humans.
  • Use Case: Ideal for software teams that want to automate the full PR review follow-up workflow, reducing manual overhead while ensuring no critical findings are missed.

Quick Start

Use the address-pr-review skill to automatically process all review findings for your current open PR, including triage, fixes, and CI monitoring, with configurable iteration and severity thresholds.

Frequently Asked Questions about address-pr-review

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

FAQPage Schema
How do I automate end-to-end GitHub pull request review workflows?

Automate GitHub pull request review workflows by orchestrating parallel review agents to collect feedback, triaging findings by severity, applying automated fixes, and monitoring CI status with configurable retry limits.

What is the best way to triage code review findings from multiple sources?

Triage code review findings by cross-validating issues across multiple parallel review sources, applying session-aware severity adjustments, and filtering out false positives to ensure only critical issues are prioritized for fixes.

How does automated CI failure recovery work for pull requests?

Automated CI failure recovery monitors CI status after applying fixes, runs local validation, and automatically retries failed CI jobs up to a configurable iteration limit before escalating remaining failures to human reviewers.

Can I run multiple code review agents in parallel on a single pull request?

Run multiple code review agents in parallel on a single pull request by spawning parallel subagents that collect comprehensive feedback across different review methodologies in one pass, reducing manual collection overhead.

How do you prevent infinite loops during automated PR fix cycles?

Prevent infinite loops during automated PR fix cycles by setting configurable iteration and severity thresholds, applying guardrails against repetitive validation errors, and escalating to humans when retry limits are exceeded.

Do I need specific review tools to use automated pull request workflow orchestration?

Automated pull request workflow orchestration integrates with various code review tools by spawning parallel review subagents, requiring no specific dependencies to coordinate finding triage, automated fixes, and iterative CI monitoring.