pev-reviewer

Review Builder changes against Architect pitch for spec compliance and code quality.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/ddpoe/pev-agent-nexus --skill pev-reviewer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pev-reviewer
Source: https://github.com/ddpoe/pev-agent-nexus/tree/main/plugins/pev/skills/pev-reviewer
Command: npx skills add https://github.com/ddpoe/pev-agent-nexus --skill pev-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a structured, repeatable approach to reviewing code changes against an Architect pitch, capturing compliance, preservation of functionality, and quality concerns without modifying the code itself.

Core Features & Use Cases

  • Guided three-pass review workflow: Spec compliance, functionality preservation, and quality assessment
  • Artifacts-aware: verifications against the cycle manifest, builder progress, and required artifacts
  • Progress reporting: writes to the cycle manifest via cortex_update_section to persist review outcomes
  • Tooling integration: leverages cortex_check, cortex_diff, cortex_history, cortex_graph, cortex_source, and related git tools to map and verify changes

Quick Start

Initiate the review by reading the cycle manifest and running cortex_check to map affected nodes, then proceed with Pass 1.

Frequently Asked Questions about pev-reviewer

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

FAQPage Schema
How do I review Builder code changes against an Architect pitch for spec compliance?

Code review for Builder changes involves a three-pass workflow verifying spec compliance, functionality preservation, and quality. This skill applies the PEV reviewer process across cycle manifests and worktrees, using cortex tools like check and diff to assess changes against the Architect pitch.

What is the PEV reviewer workflow for cycle manifests?

The PEV reviewer workflow is a structured process for reviewing code changes against an Architect pitch. It applies a three-pass assessment covering spec compliance, functionality preservation, and quality, persisting review outcomes directly to the cycle manifest via cortex_update_section.

Do I need cortex tooling to run code reviews on cycle manifests and worktrees?

Yes, cortex tooling is required to run these code reviews. The skill leverages cortex_check, cortex_diff, cortex_history, cortex_graph, and cortex_source to map affected nodes, verify diffs, and assess changes across worktrees and cycle manifests.

How do I persist code review outcomes to a cycle manifest?

To persist code review outcomes to a cycle manifest, the skill uses cortex_update_section to write progress and results. This ensures that spec compliance, functionality preservation, and quality assessments are saved as artifacts within the manifest.

Can I use this approach to verify functionality preservation without modifying the code?

Yes, you can verify functionality preservation without modifying the code. The skill acts as a read-only reviewer, applying a three-pass workflow to assess spec compliance, functionality, and quality against the Architect pitch using cortex diff and history tools.

What's the best way to start a guided code review using cortex tools?

The best way to start a guided code review is to read the cycle manifest and run cortex_check to map affected nodes. From there, proceed with the first pass of the three-pass workflow to verify spec compliance against the Architect pitch.