web-data-visualization
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the use of third-party visualization libraries (Chart.js, Apache ECharts, and D3.js) from well-known and reputable sources including jsDelivr CDN and d3js.org. These are standard resources for web development and do not pose an inherent security risk in this context.
- [DATA_EXFILTRATION]: No evidence of data exfiltration was found. Network operations are limited to fetching a local configuration file (
/web/_design/data.json) and loading public libraries from established CDNs. - [PROMPT_INJECTION]: The instructions do not contain any patterns typical of prompt injection, such as attempts to bypass safety filters, override system instructions, or extract system prompts.
- [COMMAND_EXECUTION]: There are no shell commands or scripts that execute arbitrary code. The workflow focuses on generating client-side JavaScript for web-based data visualization.
- [INDIRECT_PROMPT_INJECTION]: The skill includes an ingestion surface where data is read from
web/_design/data.json. While this represents a potential surface for indirect injection if the source data were maliciously crafted, the skill's primary purpose is data mapping to chart components which limits the interpretative autonomy of the agent regarding that data. - Ingestion points:
web/_design/data.json - Boundary markers: Not explicitly defined in the prompt interpolation logic.
- Capability inventory: No subprocess calls, file-writes, or dangerous network operations detected.
- Sanitization: Standard library rendering (Canvas/SVG) implicitly handles data as literal values, reducing the risk of instructional execution.
Audit Metadata