skills.rest
hamzaPixl
hamzaPixlCOMMUNITY
Explore all skills by this author →
Website𝕏@pixlHamzaGitHub

cross-review

Run dual-model code reviews on pull request diffs and match findings by file and line.

2|Updated Mar 16, 2026
Search Tags:#code review#claude#consensus#diff#cross-model#pull-request
One-click install
npx skills add https://github.com/hamzaPixl/pixl-ai --skill cross-review-hamzapixl
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cross-review
Source: https://github.com/hamzaPixl/pixl-ai/tree/main/packages/crew/skills/cross-review
Command: npx skills add https://github.com/hamzaPixl/pixl-ai --skill cross-review-hamzapixl

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides independent, parallel reviews of the same code diff using two different model tiers to surface high-confidence issues, reduce single-model blindspots, and highlight disagreements that require human judgment.

Core Features & Use Cases

  • Dual independent reviews: Runs two Claude tiers (Sonnet + Haiku) on the identical diff to obtain separate findings.
  • Consensus building: Matches findings by file and line, boosts confidence for consensus items, and classifies primary-only, secondary-only, and disagreements.
  • Use Case: Best used for critical pull requests affecting authentication, payments, data integrity, or public API contracts where cross-validation increases confidence before merging.

Quick Start

Run a cross-review on the target pull request or branch to generate a consensus table and flagged disagreements for decision-making.

Frequently Asked Questions about cross-review

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

FAQPage Schema
How do I run a cross-model code review on a pull request diff?▼

You can run a cross-model code review by applying the Skill directly to pull request diffs or branch comparisons, which triggers two independent Claude model tiers to analyze the identical diff and output a consensus table.

What is cross-model consensus building in code review?▼

Cross-model consensus building in code review matches findings from two independent AI models by file and line, boosting confidence for overlapping issues and classifying primary-only, secondary-only, and disagreements for human review.

When should I use dual-model code reviews for pull requests?▼

Use dual-model code reviews for critical pull requests affecting authentication, payments, data integrity, or public API contracts, where cross-validation increases confidence and reduces single-model blindspots before merging.

Can I use Claude models to find disagreements in code diffs?▼

Yes, you can use Claude models to find disagreements in code diffs by running two separate tiers on the identical diff to compare findings, flag conflicts, and highlight areas requiring human judgment.

What are the limitations of cross-model code reviews?▼

Cross-model code reviews are limited by their reliance on two distinct Claude tiers for analysis, meaning they require sufficient context for both models and still depend on human review to resolve flagged disagreements and final merge decisions.

Related Skills

Explore More Software Engineering Skills→
Analog-LabsAnalog-Labs

codex

Review code diffs against the base branch using the Codex CLI.

Official
Advanced
dralgorhythmdralgorhythm
111

reaching-consensus

Facilitate group decision-making using Fist of Five and Gradients of Agreement.

Community
Intermediate
dzineerdzineer

update-2-git-with-diff

Generate a Git diff, create a pull request, and merge into the main branch.

Community
Intermediate
hirokimryhirokimry
1

review-loop

Iteratively review, validate, and repair code until findings are resolved.

Community
Advanced