build-custom-chart
Pass
Audited by Gen Agent Trust Hub on Nov 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process user-provided chart requirements and Vega/Vega-Lite specifications to generate AML (Holistics) chart files. This ingestion of external data creates a potential surface for indirect prompt injection where malicious input could attempt to influence the generated code. However, the risk is mitigated by the structured nature of the AML schema and the platform's context-aware interpolation.
- Ingestion points: User descriptions and external chart specifications (Vega-Lite) are used as inputs for code generation in
SKILL.md. - Boundary markers: The skill specifies the use of
@{...}placeholders for runtime binding, which the platform handles via context-aware substitution. - Capability inventory: The skill's primary capability is writing
.chart.amlfiles to the.holistics/library/custom_charts/directory. - Sanitization: The instructions rely on the Holistics platform's internal rendering engine to handle quoting and sanitization of interpolated values.
Audit Metadata