The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of shell commands to manage the Django environment, including database migrations, test execution, and security scanning tools across multiple files and phases.
[CREDENTIALS_UNSAFE]: Phase 6 contains a shell command template that creates a Django superuser with hardcoded administrative credentials ('admin@example.com', 'admin'). While documented for test purposes, these credentials could be inadvertently committed or used in sensitive environments.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it processes and reports on untrusted data from external sources.
Ingestion points: Processes output from git diff, pip-audit, and bandit-report.json to generate verification reports.
Boundary markers: Absent; there are no delimiters or instructions provided to the agent to disregard potential instructions embedded in the processed file content or tool outputs.
Capability inventory: The skill has access to the local shell for command execution (python manage.py, pytest), file system access for reading code and writing reports (bandit-report.json), and database interaction capabilities.
Sanitization: No sanitization or validation of the ingested external content is performed before it is presented to the agent for analysis.

django-verification