health-trend-analyzer
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The interactive HTML reports utilize styling and visualization libraries (Tailwind CSS, Lucide Icons, and ECharts) referenced from well-known technology services via standard CDNs (JSDelivr and Unpkg). These resources are used solely for the presentation of the report and are documented and safe.
- [PROMPT_INJECTION]: The skill ingests user-controlled health data from local JSON logs to populate its reports.
- Ingestion points: Reads health records from various JSON files in the data directory (e.g., data/symptoms/, data/mood/).
- Boundary markers: Data is structured in JSON but contains free-text values.
- Capability inventory: Uses the Write tool to generate local HTML reports.
- Sanitization: There are no explicit instructions for the agent to sanitize or escape user-controlled strings before they are written into the report. However, because the process is local and uses the user's own data, the risk is negligible and inherent to the reporting functionality.
Audit Metadata