The Agent Skills Directory

[SAFE]: The skill serves as a security documentation resource for Django developers. It contains no malicious instructions, obfuscation, or data exfiltration patterns.
[COMMAND_EXECUTION]: All code snippets provided are standard Python/Django patterns for application logic. There are no calls to subprocesses or shell commands that pose a risk to the environment.
[CREDENTIALS_UNSAFE]: The skill correctly instructs users to manage secrets like SECRET_KEY and DATABASE_URL via environment variables. Example strings provided in .env snippets are placeholders for educational purposes and are explicitly labeled as such.
[SQL_INJECTION]: The skill includes a dedicated section on preventing SQL injection, demonstrating the safe use of the Django ORM and parameterized raw queries. It correctly identifies string interpolation in raw SQL as a vulnerability.
[PROMPT_INJECTION]: There are no instructions that attempt to override agent behavior or bypass safety filters. The language is purely instructional and follows a standard technical guide format.
[EXTERNAL_DOWNLOADS]: The skill mentions common Python packages for security (e.g., python-decouple, django-environ) but does not attempt to download or execute external scripts at runtime.

django-security