The Agent Skills Directory

[SAFE]: The skill acts as a text-processing assistant for prompt optimization. It does not request any tool permissions (allowed-tools is empty) and does not perform any autonomous file system or network operations.
[EXTERNAL_DOWNLOADS]: The README provides standard manual installation instructions using git clone from a public GitHub repository. No automated, hidden, or piped remote code execution patterns (e.g., curl|bash) are present in the skill's instructions or scripts.
[PROMPT_INJECTION]: The skill uses instructional 'Primacy Zone' rules to maintain its persona as a prompt engineer. These instructions are standard for defining agent behavior and do not contain attempts to bypass underlying model safety guidelines or extract system prompts.
[DATA_EXFILTRATION]: No patterns for accessing sensitive files (e.g., .ssh, .aws, .env) or hardcoded credentials were found. The skill does not possess the capability to send data to external servers.

prompt-master