The Agent Skills Directory

[SAFE]: The skill utilizes the Read tool to access local reference files from structured paths. This is essential for its function and does not involve unauthorized file access.
[SAFE]: Instructions are centered on persona consistency and logical reasoning, with no evidence of prompt injection or attempts to bypass agent safety constraints.
[SAFE]: The skill does not perform network operations, download external scripts, or execute shell commands.
[SAFE]: No sensitive information, such as credentials or private keys, is exposed or accessed by the skill.

dialogue