claude-code

BENIGN with notable operational risk. The skill is coherent with its stated purpose and uses official Anthropic distribution and endpoints, but it delegates broad coding actions to an external AI CLI and encourages fully automated execution with permission skipping, which materially increases security risk in real use.