test-and-fix

To fix failing tests automatically, run the test-and-fix process which detects your test runner, executes the suite, and applies targeted source-code corrections up to three times. It prioritizes root-cause changes over modifying test assertions to ensure rapid recovery.

Yes, automatic test fixing works with Python and Go projects by detecting configuration files like pyproject.toml or go.mod. It also supports JavaScript/Node and Rust environments using package.json or Cargo.toml to determine the correct test runner.

An auto-fix loop handles multiple failures by running tests, parsing failure output, and applying targeted fixes to the source code. It repeats this cycle up to three iterations, stopping when all tests pass or the maximum attempts are reached.

The best way to debug failing CI test suites without modifying tests is using an auto-fix loop that prioritizes root-cause changes in the source code. It parses failure output and applies corrections directly to the codebase, leaving test assertions intact.

If automated test fixes exceed three iterations, the auto-fix loop stops running. The tool reports what was changed after each iteration so you can review progress and manually resolve any remaining failing tests.

The auto-fix tool modifies source code instead of test files to address the root cause of failures directly. By prioritizing source-code corrections over altering test assertions, it maintains test integrity and ensures the application logic functions correctly.