The Agent Skills Directory

[No Code] (SAFE): The skill consists entirely of markdown instructions and scoring logic without any associated scripts or executable files.
[Data Exposure] (SAFE): The skill references a specific local path ~/.claude/skills/x-create/references/user-profile.md to personalize its scoring. This is an application-specific configuration file rather than a sensitive system credential or private key.
[Indirect Prompt Injection] (LOW): The skill processes external materials (raw text, URLs) which could potentially contain hidden instructions. However, since the skill has no destructive capabilities (no file writing, no network calls, no command execution), the risk of exploitation is negligible. It merely outputs a formatted text report.

x-filter