The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and YAML configuration metadata. No scripts, binaries, or executable files are included in the package, which significantly limits the potential for unauthorized or malicious code execution.
[SAFE]: The instructions include an explicit security directive to redact secrets, credentials, and private contact information from generated handoff documents. This mitigation reduces the risk of sensitive data exposure when transferring state between agents or sessions.
[SAFE]: Recommended file operations are restricted to legitimate locations such as the operating system's temporary directory or user-approved project folders, aligning with standard practices for state persistence and ensuring user oversight of created artifacts.

handoff