data-visualization-expert
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, hidden code, or instructions to bypass safety guidelines were found within the skill file. The skill's design and configuration are entirely consistent with its stated purpose.- [COMMAND_EXECUTION]: The skill is authorized to use shell commands to execute Python or R scripts it generates at runtime. This behavior is necessary for the creation of visualization artifacts and is intended for local execution within the agent's environment.- [PROMPT_INJECTION]: The skill's workflow involves reading data from external files like CSV and JSON, which serves as an interface for indirect prompt injection. However, no evidence of malicious instructions or exploitable patterns was detected in the skill's logic.
Audit Metadata