pattern-detection
Detect and document existing code patterns in a codebase. Use to maintain consistency with established conventions when writing new code.
When & Why to Use This Skill
The Pattern Detection skill automates the identification and documentation of established coding conventions within a codebase. By analyzing naming styles, error handling, and testing structures, it ensures architectural consistency and provides a quantitative consistency score to guide refactoring and standardization efforts, making it an essential tool for maintaining high-quality, idiomatic code.
Use Cases
- Developer Onboarding: Helping new team members quickly grasp and adopt project-specific coding styles and conventions to ensure their contributions match the existing codebase.
- Technical Debt Assessment: Identifying deviations from established patterns in legacy code to prioritize refactoring tasks and improve overall system maintainability.
- Automated Consistency Audits: Generating detailed reports and scores for naming, error handling, and testing patterns to maintain architectural integrity across large-scale projects.
- AI-Assisted Development: Providing structural context to AI agents so they can generate new code that seamlessly integrates with the established patterns of a specific repository.
| name | pattern-detection |
|---|---|
| description | Detect and document existing code patterns in a codebase. Use to |
Pattern Detection Skill
Purpose
Identify and document patterns for consistency enforcement.
Pattern Categories
1. Naming Conventions
Reference: patterns/naming-conventions.md
Detection method:
- Sample 10+ files across different directories
- Extract function/class/variable names
- Identify dominant patterns
2. Error Handling
Reference: patterns/error-handling.md
Detection method:
- Search for try/catch blocks
- Find custom error classes
- Check for error middleware
3. Testing Patterns
Reference: patterns/testing-patterns.md
Detection method:
- Locate test files
- Analyze test structure
- Identify mocking approach
Consistency Scoring
Rate codebase consistency:
- High (90%+): Strong patterns, few deviations
- Medium (70-89%): Clear patterns with some variation
- Low (<70%): Inconsistent, needs standardization
Output Format
Create patterns document with:
- Detected patterns with examples
- Consistency score per category
- Deviation examples
- Recommendations for standardization
Storage Location
Save to: docs/research/patterns-{date}.md