The Agent Skills Directory

[NO_CODE] (SAFE): The skill consists exclusively of Markdown documentation files. It does not contain any executable scripts, binaries, or configuration files that would initiate logic or modify system state.- [COMMAND_EXECUTION] (INFO): While the documentation contains numerous code snippets demonstrating Remote Code Execution (RCE) and command injection (e.g., pickle.loads, unserialize, os.system, eval), these are explicitly labeled as "DANGEROUS" examples for educational reference. They are not part of an executable framework and do not pose a risk unless an agent is specifically instructed to execute the text from these files.- [DATA_EXFILTRATION] (INFO): The reference material mentions sensitive file paths such as /etc/passwd, /etc/shadow, and ~/.ssh/id_rsa to illustrate path traversal and data exposure vulnerabilities. These are technical descriptions of risks, not active attempts to access or exfiltrate data.

sharp-edges