The Agent Skills Directory

[SAFE]: The skill instructions and the accompanying implementation playbook do not contain any malicious instructions, prompt injection patterns, or obfuscated content. All code examples provided are for educational and refactoring purposes, illustrating standard design patterns (SOLID, Factory, Strategy, etc.) and common code smells.
[DATA_EXFILTRATION]: No exfiltration patterns or unauthorized network operations were detected. The skill examples use standard, benign domains such as 'smtp.gmail.com' and 'localhost' for illustrative purposes within refactoring scenarios.
[REMOTE_CODE_EXECUTION]: No remote code execution or untrusted dependency installation patterns were found. The skill includes examples of GitHub Actions and tool configurations (Ruff, ESLint, SonarQube) which are presented as documentation for the user to implement, rather than commands for the agent to execute at runtime.
[CREDENTIALS_UNSAFE]: No hardcoded secrets or credentials were found. Examples of database connections in the 'monolith' code smell illustration use placeholder values like 'user=root' and 'host=localhost' which are standard for documentation and do not represent a security risk.
[PROMPT_INJECTION]: The skill uses natural instructional language and does not attempt to bypass agent safety filters or override system instructions.
[COMMAND_EXECUTION]: No dangerous shell commands or privilege escalation attempts were detected. The skill focuses on code analysis and transformation through standard refactoring practices.

clean-code-refactoring