The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional text to guide the agent's output formatting. It contains no executable scripts, commands, or network operations.
[SAFE]: The skill includes explicit safety instructions to override formatting rules in favor of safety, such as requiring user confirmation before executing potentially destructive actions.
[SAFE]: All examples provided in the instructions (such as error message formatting) are purely illustrative and do not contain hardcoded credentials or sensitive data access patterns.
[SAFE]: The skill is designed for one-shot invocation and explicitly prohibits persistent state, which minimizes the risk of unintended behavior in subsequent conversation turns.

tk-adhd