- Home
- Skills
- Ntcoding
- Claude Skillz
- Lightweight Design Analysis
lightweight-design-analysis_skill
- Python
247
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill ntcoding/claude-skillz --skill lightweight-design-analysis- README.md5.0 KB
- SKILL.md15.8 KB
Overview
This skill analyzes code for design quality improvements across eight rigorous dimensions: Naming, Object Calisthenics, Coupling & Cohesion, Immutability, Domain Integrity, Type System, Simplicity, and Performance. It enforces an evidence-based protocol that first understands code flow, then evaluates each dimension with concrete file:line findings and actionable recommendations. Output is focused, reproducible, and includes code snippets and severity levels.
How this skill works
The skill auto-invokes an implementation-analysis protocol to build a precise model of code flow, responsibilities, and relationships with file:line references. It then inspects the code systematically across the eight dimensions, detecting concrete violations (e.g., generic names, nested indentation, feature envy, mutable state, anemic domain logic, poor types, duplication, or algorithmic inefficiency). For every finding it emits severity (🔴/🟡), exact locations, code examples, and suggested before/after fixes.
When to use it
- Requesting a design-level code review of a class or small module
- Looking for refactoring opportunities focused on maintainability
- Validating domain model integrity and invariants
- Assessing coupling, cohesion, and type-safety before release
- Investigating suspected performance hotspots with evidence
Best practices
- Always run the implementation-analysis step first to gather file:line flow before asserting issues
- Report every finding with explicit file:line and a code snippet demonstrating the problem
- Prioritize fixes with clear impact: protect invariants, reduce coupling, prefer immutability
- Aim for single-responsibility units and extract methods when indentation or complexity rises
- Prefer explicit domain types and readonly state to make illegal states unrepresentable
- Only flag performance issues when there is clear evidence and a meaningful payoff
Example use cases
- Analyze a UserService or Order module for feature envy and misplaced business logic
- Identify naming and primitive-obession issues in utility-heavy modules
- Detect mutable state and array mutations in a Cart/Checkout implementation
- Find nested logic violating Object Calisthenics and propose extraction points
- Locate O(n²) patterns (find/filter inside loops) and suggest Map-based refactors
FAQ
Small-scale analysis: a single class, module, or a small set of related files. It is not intended for whole-repo architecture audits.
Why must implementation analysis run first?
Understanding control flow, dependencies, and responsibilities is required to make evidence-based findings and avoid false positives.
How are severities determined?
Severity is based on concrete impact: safety/invariant violations, high coupling, or O(n²) hotspots are 🔴; naming, duplication, and mutability suggestions are 🟡.