The Agent Skills Directory

[SAFE]: No security risks were identified in this skill. The content is limited to instructional prose and does not include shell scripts, binary files, or network-enabled commands.
[PROMPT_INJECTION]: The skill uses a metaphorical 'spell' theme but does not attempt to override agent safety protocols or ignore system instructions. The instructions remain within the bounds of standard investigative tasks.
[DATA_EXFILTRATION]: There are no commands or instructions that facilitate the movement of data to external or untrusted domains. The search process described is limited to observable internal signals like commit history and org charts.

locate-creature