The Agent Skills Directory

[SAFE]: The skill performs network operations to flows-api.jetty.io to check account status and register tasks. As the skill is authored by jettyio, this is considered legitimate communication with vendor infrastructure.
[SAFE]: The skill accesses ~/.config/jetty/token to authenticate with its own API. This is standard practice for platform-integrated tools for managing configuration and authentication.
[SAFE]: A temporary shell script is generated in /tmp to perform structural validation of the created runbook. The script's logic is restricted to local file parsing and validation.
[SAFE]: The skill processes user-supplied task descriptions to populate templates. This is the primary function of the wizard and does not involve unsafe execution of the input.

create-runbook