← Back to Docs
Recipe

Mutation Testing

Mutation testing measures the quality of your test suite by injecting small faults into your source code and verifying that your tests catch them. If a mutant survives, your tests have a gap.

Core Idea

A mutation operator makes a tiny change — flipping a boolean, removing a condition, swapping an operator. The mutated code is called a mutant. Your test suite runs against each mutant. If a test fails, the mutant is killed. If all tests pass, the mutant survived — and you have uncovered a testing blind spot.

Common Operators

  • AOR — Arithmetic operator replacement ( + → - , * → / )
  • ROR — Relational operator replacement ( > → >= , == → != )
  • COR — Conditional operator replacement ( && → || )
  • SBR — Statement block removal (delete entire if-body)

Mutation Score

The ratio of killed mutants to total mutants. A score below 80% signals weak tests. Aim for 100% — every surviving mutant represents untested behavior that can ship as a bug.

Quick Start

  1. Install a mutation runner (Stryker, PIT, mutmut).
  2. Point it at your source and test directories.
  3. Run a baseline to generate mutants.
  4. Inspect survivors — each one is a missing assertion.
  5. Write the test that kills it, then repeat.

Next: Property-Based Testing →