The Agent Skills Directory

[SAFE]: The skill is entirely informational, providing architectural guidance on terminal output streams, exit codes, interactivity, and accessibility.
[SAFE]: Contains security best practices, specifically advising developers to treat machine-readable output formats as trust boundaries and to implement proper quoting/escaping to prevent parameter injection in downstream consumers.
[NO_CODE]: The skill does not provide or execute any scripts, binaries, or automated shell commands. Examples provided (such as shell completion) are for human reference and manual implementation by a developer.

terminal-ux