The Agent Skills Directory

[Prompt Injection] (SAFE): The skill contains no instructions designed to bypass safety filters or override agent behavior. The content is purely educational and logic-based.
[Data Exposure & Exfiltration] (SAFE): No access to sensitive file paths (~/.ssh, etc.) or network exfiltration patterns were found. The script operates entirely on local, hardcoded data.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): Although the documentation and scripts mention installation commands (e.g., 'pip install pyrit'), these are provided as informational metadata for the user. The skill itself does not execute these commands or download remote scripts.
[Command Execution] (SAFE): The Python script framework-selector.py performs simple string matching and logic to provide recommendations. It does not use os.system, subprocess, or eval to execute code or system commands.
[Obfuscation] (SAFE): All files are written in clear text with no evidence of Base64, zero-width characters, or other encoding techniques used to hide malicious intent.

red-team-frameworks