Code Complexity Analyzer

Cyclomatic complexity (CC) measures the number of independent paths through a function — essentially how many decision points (if, else if, case, loops, ternary, logical operators) exist. A CC of 1-5 means the function is simple and easy to test. 6-10 is moderate and may benefit from refactoring. Above 10 is complex and hard to test thoroughly — each path needs at least one test case. Above 20 is very high risk and should be split into smaller functions. This is especially important for AI-generated code, which often produces long functions with many branches.

Cognitive complexity measures how hard code is for a human to understand, as defined by SonarSource. Unlike cyclomatic complexity which counts all paths equally, cognitive complexity adds nesting penalties — an if inside a for inside another if scores much higher than three sequential if statements. This better reflects the mental effort needed to follow the code. A function with CC=10 might have cognitive complexity of 5 (flat structure) or 25 (deeply nested). DevBolt shows both metrics so you can identify functions that are both hard to test and hard to read.

The maintainability index (MI) is a composite metric originally developed at Carnegie Mellon and adopted by Microsoft Visual Studio. It combines cyclomatic complexity, lines of code, and Halstead volume (a measure of code vocabulary) into a single 0-100 score. Higher is better: 80-100 is grade A (highly maintainable), 60-79 is grade B, 40-59 is grade C, 20-39 is grade D, and below 20 is grade F. DevBolt uses a simplified version of this formula optimized for JavaScript and TypeScript functions.

DevBolt's Code Complexity Analyzer uses pattern-based static analysis to identify functions and calculate metrics. It handles standard JavaScript and TypeScript function declarations, arrow functions, class methods, and common patterns. The metrics are approximations — for precise results in CI/CD pipelines, use ESLint's complexity rule, SonarQube, or CodeClimate. This tool is ideal for quick code reviews, evaluating AI-generated code quality, and learning about complexity metrics without installing anything.