The Agent Skills Directory

[SAFE]: The skill functions as a router, delegating tasks to child skills located in the parent directory. This structure promotes modularity and does not introduce security risks.
[SAFE]: The Python script scripts/check_personalization.py performs local validation of the profile.yaml file and the existence of skill directories. It uses standard library modules and performs no network or unsafe code execution.
[SAFE]: The bash hook hooks/pre-tool.sh is used for logging and providing helpful reminders during tool usage. It does not perform any privileged or dangerous operations.
[SAFE]: The skill configuration in .skillmeta.yml implements a security-positive policy by explicitly forbidding the contribution of local personalization files (which might contain sensitive data) to upstream repositories.

personalize