x-card
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation describes fetching component catalogs from remote URLs (e.g., a2ui.org). This is a standard mechanism for retrieving UI schemas in the A2UI protocol.\n- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of structured data from external sources and user interactions, creating a potential surface for indirect injection.\n
- Ingestion points: Command streams passed to the XCard.Box component in SKILL.md and user input mapped via JSON Pointers.\n
- Boundary markers: Data is processed through a declarative JSON protocol rather than direct script execution, which acts as a structural boundary.\n
- Capability inventory: Support for opening URLs via built-in functions (ACTIONS.md) and fetching remote JSON schemas via loadCatalog (CATALOG.md).\n
- Sanitization: Component properties are validated against registered catalogs to ensure only permitted components and prop types are rendered.
Audit Metadata