The Agent Skills Directory

[SAFE]: The skill performs standard local operations, including reading codebase files for context during its context gathering phase and writing documentation to the project's docs/plans/ directory.
[SAFE]: No external network operations, remote code downloads, or unauthorized command executions were detected.
[SAFE]: The skill uses AskUserQuestion to interact with the user and does not attempt to bypass any safety guidelines or override agent behavior.
[SAFE]: There is no evidence of obfuscation, credential harvesting, or persistence mechanisms.

interview