data-to-ui
Fail
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: CRITICALPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill documentation defines workflows for reading and processing external JSON schemas and data structures, which is an ingestion point for indirect prompt injection. \n
- Ingestion points: JSON schemas and data structures ingested in Workflow 1. \n
- Boundary markers: The patterns do not specify the use of delimiters or 'ignore' instructions for external data. \n
- Capability inventory: The skill utilizes data transformation and UI rendering; it does not include dangerous capabilities like shell execution or arbitrary file writes. \n
- Sanitization: While the skill emphasizes type safety (using 'unknown' and type guards), it does not explicitly instruct on sanitizing natural language content to prevent instruction injection. \n- Malicious URL Detection (SAFE): An automated scanner flagged 'now.ge' as a phishing site. Analysis confirms this is a false positive; the string is a literal substring of the standard JavaScript 'now.getTime()' method used within the formatting utilities section. No network requests to this domain are present in the code.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata