skill-improve

Skill improvement through test-fix cycles is automated by running iterative loops that test, diagnose, patch, and revalidate skill definitions to enforce quality and consistency across a repository.

A baseline test for skill definitions runs static checks to collect failures and warnings, establishing a quality reference point before diagnosing issues and proposing targeted rewrites.

Diagnosing and fixing static check failures involves running baseline tests, identifying issues across multiple checks, and applying minimal targeted rewrites to resolve the detected quality problems.

Yes, retesting evaluates changes across static and category checks, and an optional reversion step allows you to revert proposed fixes if the updated quality scores do not improve.

A multi-phase workflow for quality assurance is executed by parsing arguments, running baseline tests, diagnosing issues, implementing fixed proposals, and optionally reverting changes based on score outcomes.

No external dependencies are required to run automated retests on skill definitions, as the improvement loop directly parses arguments and applies static and category checks internally.