The Agent Skills Directory

[SAFE]: The skill implements strong credential management by storing sensitive data in a local .elastic-credentials file with restricted owner-only permissions (mode 0600) and utilizing .env files for configuration.\n- [SAFE]: The Python script provides protection against command injection when loading credentials via shell eval by using shlex.quote for values and validating environment variable names against a strict regex pattern (^[A-Za-z_][A-Za-z0-9_]*$).\n- [SAFE]: The instructions include a "Critical principles" section that explicitly forbids the agent from revealing API keys, passwords, or the contents of the credential file in the chat interface or thinking traces.\n- [SAFE]: Network operations are directed solely to the official vendor API at api.elastic-cloud.com, and the skill incorporates confirmation steps for destructive actions like deleting projects or resetting credentials.

cloud-manage-project