The Agent Skills Directory

[SAFE]: The skill incorporates comprehensive security constraints, explicitly prohibiting the storage of secrets in settings files and requiring the use of environment variables. It also enforces the use of Django's built-in security features and parameterization to prevent SQL injection.
[INDIRECT_PROMPT_INJECTION]: The skill processes project source code (models.py, settings.py) as ingestion points. While this presents an attack surface for data-driven instructions, the risk is minimized as the skill focuses on standard code generation and local development tasks. Ingestion points include local project files; Boundary markers are not explicitly defined; Capabilities include standard file-write and manage.py command execution; Sanitization is addressed through mandatory DRF serializer validation logic.
[COMMAND_EXECUTION]: The core workflow involves standard Django management commands (makemigrations, migrate) and verification via curl. These are routine operations for the intended use case and are not associated with malicious intent.

django-expert