The Agent Skills Directory

[COMMAND_EXECUTION]: The helper script scripts/skate-env.py interacts with the local environment by executing git and skate commands. It uses list-based arguments for all subprocess.run calls, which prevents shell injection attacks. The script is used to derive repository roots and manage the secret database.
[SAFE]: The skill follows established security best practices for secret management. It instructs agents to store sensitive information in a dedicated, local key-value store rather than committing it to version control or exposing it in conversation logs. The script logic for repo-scoping databases is robust and prevents cross-repository data contamination.

skate