cargo-crap: Finding Untested Complexity in AI-Generated Rust Code (minikin.me)

A new Rust tool, cargo-crap, has been developed to help identify untested complexity in AI-generated Rust code. It creates a "Change Risk Anti-Patterns" (CRAP) metric that allows developers to assess whether a function is safe to modify, despite compiling correctly. This tool combines two crucial factors: cyclomatic complexity, which measures the number of paths through a function, and test coverage to identify complex yet poorly tested code. With many AI agents now involved in coding, this tool serves as a vital guardrail, alerting developers to potential risks introduced by automated code changes that might increase complexity without adequate testing. The CRAP metric is based on a formula that quantifies the interplay between complexity and coverage, allowing developers to focus on high-risk areas requiring attention. By integrating cargo-crap into their development workflow, teams can quickly pinpoint functions that need refactoring or additional testing. It not only facilitates a more informed code review process but also contributes to maintaining long-term code quality, especially in environments where AI algorithms are increasingly generating and altering code. This ensures that as developers accelerate their coding practices, they don't overlook the risks associated with modifying complex and untested code segments.