data-storytelling-analyst
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions and supporting documentation focus entirely on legitimate data science and information design practices. It recommends industry-standard libraries such as Matplotlib, Seaborn, and Plotly, and follows established design and accessibility standards without requesting unnecessary permissions or performing risky operations.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided datasets for analysis and visualization. This creates a potential surface for indirect prompt injection where malicious instructions could be embedded within data files (e.g., CSV or JSON) to influence the agent's output.
- Ingestion points: User-supplied datasets identified during the exploratory data analysis (EDA) phase (SKILL.md).
- Boundary markers: No explicit delimiters or boundary warnings are included in the instructions for handling data content.
- Capability inventory: The agent has the capability to generate and execute Python or R code for data processing and visualization.
- Sanitization: The skill recommends standard data cleaning (handling missing values, type conversion) but does not include specific security-focused sanitization for text-based data entries.
Audit Metadata