unfold-admin
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates a modern administrative interface that processes and displays untrusted data from Django models, which could be leveraged for indirect prompt injection if the agent reads and acts upon data containing hidden instructions.
- Ingestion points: Data from Django models is ingested and displayed through ModelAdmin and dataset configurations (SKILL.md, references/dashboard.md).
- Boundary markers: The instructions do not define delimiters or markers to distinguish between legitimate model content and instructions meant for the agent.
- Capability inventory: The skill enables significant capabilities including global actions, row-level operations, and custom form submissions across all reference files.
- Sanitization: While Django provides default escaping, this skill explicitly supports rendering unescaped HTML content through features like the 'html' preprocessor and the 'WysiwygWidget' (references/widgets-inlines.md).
Audit Metadata