test-runner

SKILL.md

For detailed test case examples for skills, agents, commands, and hooks, see examples.md.

For a catalog of common failure patterns by customization type, see common-failures.md.

In read-only mode (default), no customizations are actually invoked - the skill analyzes configurations and documentation to assess expected behavior. In active mode, skills are invoked with test queries to verify actual behavior.

Ejecuta pruebas sistemáticas sobre las personalizaciones de Claude Code. Ejecuta consultas de muestra, valida respuestas, genera informes de prueba e identifica casos extremos para agentes, comandos, habilidades y enlaces. Fuente: philoserf/claude-code-setup.

Ver original