skills.rest
dljsjr
dljsjrCOMMUNITY
Explore all skills by this author →
GitHub

code-review

Detect code quality issues in pull requests, diffs, and codebases.

Updated Mar 20, 2026
Search Tags:#software engineering#quality assurance#static analysis#code review#security#metrics#risk assessment
One-click install
npx skills add https://github.com/dljsjr/sandpiper-ai --skill code-review-dljsjr
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/dljsjr/sandpiper-ai/tree/main/skills/sandpiper/code-review
Command: npx skills add https://github.com/dljsjr/sandpiper-ai --skill code-review-dljsjr

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Structured, evidence-based code reviews that blend quantitative tool analysis with human judgment to catch structural issues, quality gaps, and architectural risks, reducing bikeshedding and subjective judgments.

Core Features & Use Cases

  • A two-mode review framework: Diff Review for changes and Health Audit for overall code health, with language-specific references to guide reviews.
  • A stepwise workflow: understand intent, measure delta, assess correctness, evaluate architecture and data integrity, and produce an actionable health summary.
  • Triggered by PRs, pasted code, or diffs, applying across languages (TS/JS, Python, Rust, Go, Java, etc.) to improve reliability and maintainability.

Quick Start

Describe the changes you want reviewed and request a structured health assessment.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I perform an evidence-based code review on a pull request?▼

Evidence-based code review analyzes pull requests and diffs to detect correctness, security, and maintainability issues. It measures code deltas and evaluates structural risks to reduce subjective bikeshedding through quantitative metrics.

Can I assess the overall health of an existing codebase across different languages?▼

Yes, you can assess existing codebase health across languages like Python, Rust, and Java. A health audit evaluates architectural integrity and data correctness to produce an actionable summary of structural quality gaps.

What is the difference between reviewing a diff and auditing overall code health?▼

Diff review measures deltas in pull requests to catch immediate correctness and security issues, while a health audit evaluates the structural assessment and maintainability of an entire existing codebase.

Does this code review process work for TypeScript and Go projects?▼

Yes, the review framework applies across languages including TypeScript, Go, Java, and Python. It uses language-specific references to guide structural assessment and evaluate data integrity reliably.

How does static analysis evaluate architectural risks in a codebase?▼

Static analysis evaluates architectural risks by measuring code deltas and assessing structural correctness. It blends quantitative tool analysis with formal assessment to identify maintainability gaps and data integrity issues.

Why should I use structured code reviews instead of manual checks?▼

Structured reviews reduce subjective judgments and bikeshedding by blending quantitative tool analysis with human judgment. They provide a stepwise workflow to catch structural issues, security gaps, and architectural risks systematically.

Related Skills

Explore More Software Engineering Skills→
synthesisengineeringsynthesisengineering
15

synthesis-pr-review

Analyzes pull request diffs for regressions, risk, and integration readiness.

Official
Intermediate
navapbcnavapbc
6

retro

Assess project health metrics and create a remediation plan with acceptance criteria.

Official
Advanced
eliottbusinesseliottbusiness

requesting-code-review

Automate pre-commit code review with static analysis and security scanning.

Community
Advanced
Rune-kitRune-kit
82

sast

Run static analysis tools and normalize findings into unified severity levels.

Official
Advanced