The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of shell commands across all 12 phases to automate environment checks, database migrations, testing, and static asset building.
[CREDENTIALS_UNSAFE]: Phase 6 contains a management command that creates a Django superuser with a hardcoded email ('admin@example.com') and password ('admin'). While these are likely intended as placeholders for development environments, they are literal credentials embedded in the instruction scripts.
[EXTERNAL_DOWNLOADS]: The skill and its provided configuration examples recommend installing numerous external development and security tools via standard package managers, including ruff, bandit, safety, and pip-audit.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection:
Ingestion points: The skill ingests untrusted data by reading local project source files and git diffs through tools like grep, bandit, and ruff (SKILL.md).
Boundary markers: Absent. There are no explicit delimiters or instructions to ignore embedded commands within the files being analyzed.
Capability inventory: The skill has broad capabilities including shell command execution, file system access, and database interaction through the Django management interface.
Sanitization: Absent. The skill does not implement validation or escaping of the content read from files before processing or reporting.

django-verification