scientific-figure-making
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill generates Python code to process and visualize untrusted data such as plot labels and category names, which provides a surface for indirect prompt injection. * Ingestion points: Data parameters in
references/api.md(e.g.,labelsinmake_grouped_bar,x_labelsinmake_heatmap). * Boundary markers: The skill does not provide delimiters or instructions to the agent to distinguish between data and embedded instructions. * Capability inventory: The agent is tasked with generating and executing complex Python visualization scripts. * Sanitization: There is no requirement for validation or sanitization of input strings before they are incorporated into the figure generating code.
Audit Metadata