mutation-testing

Mutation testing detects weak tests by introducing intentional code changes (mutants) and checking if your tests catch them. Install mutmut, run `mutmut run` on your Python project, then review `mutmut results` to identify tests that fail to catch mutations, revealing gaps in test coverage.

Mutation testing modifies your code with operators like arithmetic, boolean, and comparison changes to verify tests validate actual behavior. By computing a mutation score—the percentage of mutants killed by tests—you measure test effectiveness and identify areas where tests pass but don't truly validate logic.

Yes, mutation testing integrates into CI/CD workflows. Use mutmut or cosmic-ray to run mutations on specific modules or files as part of your pipeline, generate mutation scores, and fail builds if scores fall below your quality threshold, enforcing test quality standards across commits.

Both tools detect weak tests through mutation testing. Mutmut is lighter and faster for unit tests; cosmic-ray offers broader mutation operators and supports integration tests. Choose based on your codebase size, test scope, and performance requirements.

Mutation testing works with unit and integration tests. It runs mutations on your code and executes test commands to measure how many mutants are caught, providing actionable recommendations to strengthen tests across both test levels.

After running mutation testing and reviewing results, fix tests that don't catch mutants by adding assertions that validate specific behavior changes. Iteratively re-run `mutmut run` and `mutmut results` until your mutation score reaches your quality target.