django-deployment
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests deployment metadata from a local project file to determine infrastructure requirements.
- Ingestion points: Reads the file .agents/django-project-context.md to infer hosting, web server, and environment strategy.
- Boundary markers: The skill does not provide delimiters or instructions to ignore potential commands within the context file.
- Capability inventory: The agent generates Python settings, Dockerfiles, and shell commands based on this untrusted input.
- Sanitization: Content from the context file is used without explicit validation.
- [COMMAND_EXECUTION]: The skill generates standard infrastructure setup commands including pip install, apt-get, and manage.py migrations.
- [SAFE]: The skill promotes secure practices by recommending environment variable usage for secrets and implementing production-grade security headers in settings templates.
Audit Metadata