The Agent Skills Directory

[SAFE]: The skill provides robust security guidance, explicitly warning against insecure practices such as custom cryptographic primitives, weak hashing algorithms (MD5/SHA-1), and IV reuse.- [COMMAND_EXECUTION]: The skill includes a 'Companion check' that uses the ls command to inspect local skill directories (e.g., ~/.claude/skills/) to identify if recommended companion skills are installed. This is a benign environment inspection used to provide relevant recommendations to the user.- [DATA_EXPOSURE]: Illustrative code examples for TLS configuration reference standard certificate paths such as /etc/ssl/private/. These are provided for educational purposes and do not represent a risk of automated credential harvesting by the skill itself.

cryptography