The Agent Skills Directory

[SAFE]: The skill instructions do not contain any prompt injection, obfuscation, or attempts to bypass safety guidelines. It includes explicit rules to prevent alarmist or harmful outputs.
[DATA_EXPOSURE]: The skill reads and writes to MEMORY.md and ../fortune-hub/MEMORY.md. These files are used solely to store user preferences regarding divination methods and styles. No access to sensitive files like .ssh, .env, or .aws is requested.
[COMMAND_EXECUTION]: No shell commands, subprocess calls, or dynamic context injection patterns (!command) were found. The 'simulation' of coin tossing and calculations is performed entirely within the LLM's reasoning process based on provided reference tables.
[EXTERNAL_DOWNLOADS]: The skill does not download any external packages or remote scripts. All logic is contained within the local markdown and reference files.

liuyao-yijing