- Home
- Skills
- Code Quality & Review
- 0-refactor-advisor
0 Refactor Advisor
Transform technical debt into a prioritized roadmap with professional-grade refactoring reports.
Free
Works with the AI Agents you already use
0-refactor-advisor
Example session with this skill installed
// userService.js
async function processUser(u, t, r, adminKey) {
if (adminKey === 'secret_key_2024') {
if (u && u.email && u.email.includes('@')) {
const db = require('./db');
const existing = await db.query(`SELECT * FROM users WHERE email='${u.email}'`);
if (existing.length === 0) {
await db.query(`INSERT INTO users (email, type, role) VALUES ('${u.email}', '${t}', '${r}')`);
console.log('user created');
// TODO: send welcome email
const db2 = require('./db');
await db2.query(`UPDATE stats SET user_count = user_count + 1 WHERE id=1`);
} else {
console.log('user exists');
}
}
}
}
- Read your context and instructions
- Compiled the 0-refactor-advisor
REFACTOR ADVISOR REPORT
════════════════════════════════════════
File/Module: userService.js — processUser()
Language: JavaScript (Node.js)
Lines: ~20
════════════════════════════════════════
TECHNICAL DEBT SCORE: 8/10
Smells found: 7 total (3 critical · 3 medium · 1 low)
QUICK WINS (each < 15 min)
────────────────────────────
- Line 2 — Magic String — extract 'secret_key_2024' to process.env.ADMIN_KEY
- Line 13 — Dead Code — remove duplicate const db2 = require('./db'); use existing db
- Line 1 — Poor Naming — rename u→user, t→type, r→role
PRIORITY REFACTORS
──────────────────────────────────────────────────────
#1 Severity: CRITICAL
Smell: SQL Injection (Magic String in query)
Location: Lines 6, 8, 10 — template literals inside db.query()
Problem: String interpolation in SQL queries is not escaped. An attacker controlling
u.email can exfiltrate or destroy the database.
Fix: Replace with parameterized queries:
db.query('SELECT * FROM users WHERE email = $1', [user.email])
[... abbreviated ...]
Connects securely to your tools. The creator never sees your data.
About this skill
High-Level Code Quality Analysis
Refactor Advisor is a specialized developer tool designed to bridge the gap between "working code" and "clean code." It acts as a deep-scanning auditor that identifies technical debt, structural flaws, and security risks using industry-standard patterns like SOLID principles and Martin Fowler’s refactoring catalog.
What it does
Unlike generic AI prompts, this skill applies a strict heuristic framework to your codebase. It categorizes issues from "Poor Naming" to "God Classes," calculates a quantitative Technical Debt Score, and identifies "Quick Wins" that provide immediate value with minimal risk.
- Detailed Auditing: Scans for 10+ specific code smells including Feature Envy, Deep Nesting, and Magic Numbers.
- Prioritized Reporting: Ranks issues by severity (Critical/Medium/Low) so you know exactly where to start.
- Risk Management: Explicitly flags when a refactor requires "Tests First" to prevent silent regressions.
- Framework Agnostic: Works across any language or environment, from legacy Node.js monoliths to modern Python microservices.
Why use this skill
Prompting a standard AI to "make my code better" often results in subjective style changes or dangerous rewrites. Refactor Advisor provides a structured report with exact line ranges and concrete transformation steps, ensuring that code quality is improved without breaking existing contracts or introducing security vulnerabilities like SQL injection.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.
Small file with no smells found (< 20 lines, clean code):: Generated code (ORM migrations, protobuf stubs, gRPC generated files): Identify the file as generated at the top of the report, mark all smell ratings as N/A, and only flag issues if it's clear that manual edits were made inside the generated section.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
8 uses
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 6 months ago
- Free with an Agensi account