skills.rest
bponghneng
bponghnengCOMMUNITY
Explore all skills by this author →
GitHub

consensus-review

Orchestrate parallel code reviews and synthesize tiered consensus reports.

Updated Dec 5, 2025
Search Tags:#code quality#automation#ci/cd#code review#pull request#ai agents
One-click install
npx skills add https://github.com/bponghneng/rouge --skill consensus-review-bponghneng
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: consensus-review
Source: https://github.com/bponghneng/rouge/tree/main/.agents/skills/consensus-review
Command: npx skills add https://github.com/bponghneng/rouge --skill consensus-review-bponghneng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of code review by leveraging multiple specialized AI agents to ensure code quality, adherence to standards, and architectural soundness before code is merged.

Core Features & Use Cases

  • Parallel Agent Review: Spawns three distinct reviewers (standards, correctness, architecture) simultaneously.
  • Consensus Synthesis: Aggregates reviewer feedback into a tiered consensus report with a quality score.
  • Plan Conformance: Optionally verifies code against a provided implementation plan.
  • Automated Posting: Posts review summaries directly to PRs/MRs.
  • Use Case: Before merging a significant feature, run this skill to get a comprehensive review that catches potential issues across different dimensions of code quality, ensuring a higher standard for all contributions.

Quick Start

Run the consensus-review skill on all local changes to get a code review.

Frequently Asked Questions about consensus-review

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

FAQPage Schema
How does multi-agent AI code review work for pull requests?▼

Automated code review with AI agents works by spawning parallel reviewers for standards, correctness, and architecture simultaneously. It synthesizes their feedback into a tiered consensus report with a quality score, which can be automatically posted to pull requests.

Can I validate code changes against an implementation plan automatically?▼

Yes, you can validate code changes against an implementation plan automatically. The review process supports optional plan file conformance checks to verify that code adheres to a provided implementation plan before merging.

How do I automate code quality checks in CI/CD before merging?▼

You automate code quality checks in CI/CD by running a multi-agent review process on local changes or pull requests. This generates a tiered consensus report with a quality score and optionally posts the review summary directly to the PR or MR.

What is the best way to review code architecture and correctness simultaneously?▼

The best way to review code architecture and correctness simultaneously is using a multi-agent approach. It spawns distinct reviewers for architecture and correctness in parallel, aggregating their feedback into a single tiered consensus report to ensure soundness.

Does multi-agent code review work without external dependencies?▼

Yes, multi-agent code review works without external dependencies. The orchestration process runs standalone to spawn parallel reviewers, synthesize consensus feedback, and post results to pull requests without requiring additional packages.

Related Skills

Explore More Software Engineering Skills→
AndriiklymiukAndriiklymiuk
35

review

Review GitHub pull requests and GitLab merge requests against repository standards.

Community
Advanced
dmzoneilldmzoneill
2

review-pr-multiagent

Automate multi-agent code review for GitLab merge requests using Claude and Gemini models.

Community
Advanced
AnveioAnveio
1

task-pull-request

Automate GitHub pull request creation, reviews, feedback, and merges via gh CLI.

Community
Intermediate
DNYoussefDNYoussef
4

when-reviewing-github-pr-use-github-code-review

Coordinate multi-agent GitHub pull request reviews with parallel security, performance, and code analysis.

Community
Advanced