The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute the hashcat binary and local bash scripts located in the scripts/hashcat/ directory to perform password cracking, benchmarking, and learning exercises.
[DYNAMIC_CONTEXT_INJECTION]: The skill utilizes the !command`` syntax in SKILL.md to perform environment checks at load time. These checks verify the installation status of hashcat (using command -v) and the presence of wrapper scripts (using test -f). These operations are benign and intended for diagnostic purposes.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of hash files and wordlists. It has a broad capability surface involving file reads and command execution. While no specific boundary markers or sanitization logic are present in the provided instructions, the risk is consistent with the standard operating procedure for password auditing tools.

hashcat