etf-analysis
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were detected. The skill provides legitimate educational and functional content for financial analysis. The Python code templates use standard data science libraries (pandas, numpy, scipy) for their intended purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates an ingestion surface for external data by providing templates to fetch financial data from the Tushare API. While financial data typically consists of structured numeric values, the processing of external content is a known architectural surface.
- Ingestion points: Data retrieval functions in
SKILL.md(e.g.,get_etf_nav,get_etf_daily). - Boundary markers: None present in the provided templates.
- Capability inventory: Data manipulation and statistical calculations using pandas and numpy. No capability for file system writes, subprocess execution, or arbitrary network requests (beyond the specific Tushare API calls) was found.
- Sanitization: Not explicitly implemented in the templates, as they are designed to handle structured financial datasets from a known provider.
- [EXTERNAL_DOWNLOADS]: The code references the Tushare API for retrieving financial market data. This is a well-known service in the quantitative finance community and is essential for the skill's stated purpose.
Audit Metadata