code-review

Coordinate code review workflows with UACRP and mpcr.

Updated Jan 11, 2026
One-click install
npx skills add https://github.com/DevGuyRash/agent-skills --skill code-review-devguyrash
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/DevGuyRash/agent-skills/tree/main/skills/code-review
Command: npx skills add https://github.com/DevGuyRash/agent-skills --skill code-review-devguyrash

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anyhow, clap, rand, serde, serde_json, time, and includes scripts (resource) and references (resource) components.

What problem does it solve?

The Code Review Skill provides a structured, two-workflow framework to perform adversarial code reviews (Reviewer) and apply review findings (Applicator) with deterministic coordination via the mpcr tool.

Core Features & Use Cases

  • Reviewer workflow: conduct adversarial reviews using UACRP, generate a formal review report, and anchor findings to code locations.
  • Applicator workflow: read completed reviews, decide whether findings are applied/declined/deferred, and track progress via initiator_status.
  • Coordination artifacts are organized under .local/reports/code_reviews/YYYY-MM-DD/ and orchestrated by the bundled mpcr CLI.
  • Optional: prebuilt mpcr binary in scripts/mpcr; supports two phase execution: ingestion and application, with references for overview files.

Quick Start

Register for a target reference as a Reviewer and begin the ingestion phase to start a review.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I coordinate formal code reviews and track their follow-up actions?

You coordinate formal code reviews and track follow-up actions using the bundled mpcr CLI, which orchestrates artifacts and statuses across Reviewer and Applicator workflows. It organizes reports under .local/reports/code_reviews/ for deterministic tracking.

What is the adversarial code review process and how does it anchor findings?

The adversarial code review process uses UACRP to conduct reviews and generate formal reports. It anchors findings directly to specific code locations, ensuring structured feedback. The Applicator workflow then tracks whether findings are applied, declined, or deferred.

How do I start a code review using the ingestion phase?

To start a code review, register for a target reference as a Reviewer and begin the ingestion phase. This initializes the mpcr orchestration tool to process proposed changes and generate the formal review artifacts.

Do I need a POSIX shell and Rust to run the code review workflow?

Yes, you need a POSIX shell to run the code review workflow. The Skill can build the bundled mpcr tool with Rust, though an optional prebuilt mpcr binary is available in scripts/mpcr to bypass compilation.

What's the best way to apply or decline findings from a completed code review?

The best way to apply or decline findings is through the Applicator workflow. It reads completed reviews, allows you to decide on findings, and tracks progress via initiator_status using the mpcr tool across two execution phases.

Can I use this code review framework for artifact management across multiple review scenarios?

Yes, you can use this framework for artifact management across typical review scenarios where changes are proposed, reviewed, and applied or declined. It coordinates deterministic artifacts and statuses using the mpcr CLI.