The Agent Skills Directory

[SAFE]: The skill performs local file system operations to compare installed agent skills with a repository clone located at ~/.turbo/repo. These operations are scoped to reading skill definitions and identifying functional differences.
[SAFE]: A dedicated 'Privacy Filter' section provides explicit instructions to the agent to identify and remove project-specific metadata, credentials, and business logic before finalizing the contribution text.
[SAFE]: The workflow incorporates mandatory human-in-the-loop checkpoints, utilizing request_user_input for the classification of changes and the final approval of the GitHub issue content.
[SAFE]: The skill interacts with external resources (the tobihagemann/turbo repository) that are owned by the skill's author, representing standard vendor-specific functionality.

contribute-turbo