code-reviewers

Coordinate two independent reviewers to challenge findings and produce a consolidated diff report.

27|2|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/maragudk/fabrik --skill code-reviewers
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-reviewers
Source: https://github.com/maragudk/fabrik/tree/main/skills/code-reviewers
Command: npx skills add https://github.com/maragudk/fabrik --skill code-reviewers

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Dispatch a team of two competing reviewers to critique a diff, challenge each other's findings, and produce a high-signal report. Use this when the user asks for a thorough code review, wants a second opinion before committing, or wants findings that have survived adversarial scrutiny. Prefer this over the solo code-review skill when rigour matters more than speed. Invoke with /code-reviewers.

Core Features & Use Cases

  • Independent reviews: two reviewers independently assess the diff to surface diverse perspectives.
  • Adversarial challenge: reviewers challenge each other's findings, strengthening trust in the final report.
  • Consolidated output: produces a four-bucket report highlighting consensus, survived challenges, contested items, and single-reviewer notes.
  • Ideal when time, rigor, and traceability matter, such as audits, open-source contributions, or critical production changes.

Quick Start

Spawn two reviewers and start an adversarial review of the current diff.

Frequently Asked Questions about code-reviewers

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

FAQPage Schema
How do I get a rigorous code review for a pull request diff?

An adversarial code review deploys two independent reviewers to critique a diff, challenge each other's findings, and output a consolidated report. This cross-checking process ensures robust validation for critical production changes.

What is an adversarial pair review workflow in agile development?

An adversarial pair review workflow involves two reviewers independently assessing a diff, entering a challenge phase to contest findings, and producing a consolidated report. This process surfaces high-signal results by filtering findings through rigorous mutual scrutiny.

When should I use a two-person code review instead of a solo review?

Use a two-person code review when rigor and traceability matter more than speed, such as for audits, open-source contributions, or critical production changes. The paired-review protocol provides cross-checked validation that solo reviewing cannot achieve.

How do I run an independent challenge and defense code review on staged changes?

Apply a paired-review protocol to staged changes where two reviewers independently analyze the diff, defend their findings against challenges, and consolidate survived items into a four-bucket report highlighting consensus and contested items.

Does an adversarial code review work on feature branches?

Yes, adversarial code review works on feature branches, staged changes, and pull requests. It coordinates a two-person workflow to rigorously validate diffs wherever cross-checking and independent reviews are needed before integration.