code-review

Review code changes using task context and six analysis lenses.

1|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/kiran-capoor94/wizard --skill code-review-kiran-capoor94
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/kiran-capoor94/wizard/tree/main/src/wizard/skills/code-review
Command: npx skills add https://github.com/kiran-capoor94/wizard --skill code-review-kiran-capoor94

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables context-aware code review by leveraging prior investigations and decisions within Wizard to improve review quality and alignment with project goals.

Core Features & Use Cases

  • Load full task context (investigations, decisions, and mental models) before reviewing changes.
  • Analyze diffs with six lenses: Correctness, Blast Radius, Invariant Violations, Observability, Tests, and Simplicity.
  • Persist review findings as notes and optionally update task status to reflect decisions.

Quick Start

Invoke code-review on a task with existing context to start a guided, context-aware review of changes.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I perform a context-aware code review on a pull request?

Context-aware code review analyzes pull request diffs by loading prior task investigations and decisions to improve alignment with project goals. It applies six review lenses including correctness, blast radius, invariant violations, observability, tests, and simplicity.

What is diff analysis and how does it use task context?

Diff analysis evaluates code changes by leveraging task context APIs to load prior investigations and mental models. This ensures the review surfaces actionable findings aligned with the original project decisions.

Can I review code changes across multiple files simultaneously?

Yes, context-aware code review supports analyzing code changes across multiple files and contexts. It loads the full task context before reviewing to ensure comprehensive coverage of all affected areas.

What review lenses are applied during code analysis?

Code analysis enforces six review lenses: Correctness, Blast Radius, Invariant Violations, Observability, Tests, and Simplicity. These lenses surface actionable findings by examining different aspects of the code changes.

How do I save code review findings and update task status?

The Skill persists review findings as notes using the save_note API and optionally updates task status via update_task. This integration ensures review decisions are recorded within the existing task context.