agentsqripts/lib/srp-violations/PROPOSED-IMPROVEMENTS.md

Comprehensive static code analysis toolkit for identifying technical debt, security vulnerabilities, performance issues, and code quality problems

# Proposed SRP Analyzer Improvements

## Option 1: Graduated Scoring
Instead of harsh penalties for 2+ functions, scale gradually:
- 1 function: 0 points (perfect)
- 2-3 functions: +1 point (minor)
- 4-6 functions: +3 points (moderate)
- 7-10 functions: +5 points (high)
- 10+ functions: +8 points (critical)

## Option 2: Context-Aware Thresholds
Different file types get different rules:
- Test files: Allow up to 10 test functions
- Utility files: Allow 3-5 related helpers
- Config files: More lenient
- Core business logic: Stricter

## Option 3: Complexity Weighting
Consider function size/complexity:
- Count lines per function
- Functions <5 lines count as 0.5
- Functions >50 lines count as 2.0
- Score based on weighted total

## Option 4: Relationship Detection
Try to detect related functions:
- Functions with similar names (openDB, closeDB)
- Functions that call each other
- Functions in same lexical scope
- Group penalty only if unrelated

## Option 5: Practical Defaults
Change the definition:
- "Strict mode": Current behavior (1 function = compliant)
- "Standard mode": Up to 3 related functions OK
- "Lenient mode": Focus on files with 5+ functions

## Recommendation
Implement Option 1 (Graduated Scoring) + Option 2 (Context-Aware) as they:
- Are easiest to implement with static analysis
- Provide more reasonable results
- Still encourage good practices
- Don't require complex heuristics