code-improve

Analyze Go, TypeScript, and Python source code for actionable improvements.

1|Updated Jan 18, 2026
One-click install
npx skills add https://github.com/knwoop/claude-code --skill code-improve
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-improve
Source: https://github.com/knwoop/claude-code/tree/main/.claude/skills/code-improve
Command: npx skills add https://github.com/knwoop/claude-code --skill code-improve

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyze source code for improvements across security, correctness, performance, refactoring, code smells, and best practices. Supports Go, TypeScript, and Python with language-specific checks.

Core Features & Use Cases

  • Multi-pass analysis with category-specific expert sub-agents.
  • Detects security, correctness, performance, refactoring, code smells, and best practices across Go/TS/Python.
  • Produces a structured report with severity levels and concrete fix suggestions.
  • Works across single files and multi-file repositories, honoring linter configurations when present.

Quick Start

Specify the files to analyze and run the multi-pass code-improvement analysis.

Frequently Asked Questions about code-improve

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

FAQPage Schema
How do I analyze source code for security and performance improvements in Go, TypeScript, and Python?▼

Multi-pass code analysis detects security, correctness, performance, and best practice issues across Go, TypeScript, and Python. It uses category-specific sub-agents to produce a structured report with severity levels and concrete fix suggestions for single files or whole repositories.

Can I scan a polyglot repository containing multiple languages like Go and TypeScript?▼

Yes, you can scan polyglot repositories containing Go, TypeScript, and Python. The analysis works across single files and multi-file repositories, applying language-specific checks and respecting linter configurations when present.

What is the best way to detect code smells and refactoring opportunities in a Python project?▼

The best way to detect code smells is through multi-pass analysis with category-specific expert sub-agents. This approach analyzes your Python source code to identify actionable refactoring opportunities while respecting existing language patterns and linter configurations.

Does the code improvement analysis respect existing linter configurations?▼

Yes, the code improvement analysis respects language patterns and linter configurations when present. It applies targeted or whole-project analysis across Go, TypeScript, and Python to ensure suggestions align with your existing project standards.

How do I get a structured report with concrete fixes for my TypeScript code?▼

You get a structured report with concrete fixes by running the multi-pass analysis on your TypeScript code. The analysis identifies actionable improvements across security and performance, outputting severity levels and specific fix suggestions.