create-catalog-item
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process user-supplied JSON Schema definitions to generate Dart code, creating an attack surface for indirect prompt injection.\n
- Ingestion points: The skill ingests untrusted schema definitions as the source for generating
CatalogItem, data class, and widget class implementations as described inSKILL.md.\n - Boundary markers: No specific boundary markers or instructions are provided to the agent to treat the schema content as potentially adversarial data or to ignore embedded instructions.\n
- Capability inventory: The resulting code utilizes
resolveContextto fetch variables from the application's data context anddispatchEventto send data via aUserActionEvent. This allows a malicious schema to potentially define actions that resolve sensitive context data (e.g., tokens or user info) and dispatch them to the application bus.\n - Sanitization: The instructions emphasize runtime safety through type casting and
try-catchblocks, but do not provide measures to sanitize the schema's metadata or property values against injection attempts aimed at the AI's logic.
Audit Metadata