The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill focuses on promoting secure development practices by providing defensive coding templates.
[COMMAND_EXECUTION]: The skill provides explicit guidance on the safe use of subprocesses, instructing developers to avoid shell=True and use argument lists to prevent shell injection vulnerabilities.
[PROMPT_INJECTION]: No evidence of prompt injection or attempts to override AI behavioral constraints was found in the instructions or templates.
[DATA_EXFILTRATION]: No network exfiltration patterns or unauthorized data access behaviors were detected. The templates manage data locally using SQLite and standard configuration directories.
[CREDENTIALS_UNSAFE]: The skill includes warnings against hardcoding secrets and provides templates for managing sensitive data using environment variables instead of hardcoded strings.

python-toolsmith