data-viz
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided raw data for visualization, which represents a potential surface for indirect prompt injection. However, the instructions include explicit guidelines for input validation to mitigate these risks.\n
- Ingestion points: Processes user-provided "raw data" for dashboards, maps, graphs, and diagrams as described in the overview and instructions (SKILL.md).\n
- Boundary markers: No specific boundary delimiters are defined in the instruction text, but the guidelines provide a logical boundary through validation requirements.\n
- Capability inventory: The skill is instructional and does not invoke subprocesses or perform network operations itself.\n
- Sanitization: Guidelines explicitly instruct the agent to "Always validate inputs before processing."\n- [SAFE]: No malicious patterns, unauthorized network activity, or obfuscation were detected in the skill instructions or metadata.
Audit Metadata