data-viz

Pass

Audited by Gen Agent Trust Hub on Oct 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill focuses on providing instructional content and static assets for data visualization. It follows best practices for accessibility and design consistency.- [INDIRECT_PROMPT_INJECTION]: The skill processes external data to generate charts and tables, which theoretically presents an indirect prompt injection surface. The risk is mitigated by the skill's requirement for rigorous data documentation (source, grain, missingness) and strict encoding rules. * Ingestion points: Processes data shapes, comparison questions, and sample data as defined in contract.json and SKILL.md. * Boundary markers: The skill requires explicit recording of data provenance and units before visualization selection. * Capability inventory: The skill generates SVG and HTML code recommendations and implementation plans. * Sanitization: The skill emphasizes 'honest encoding' and 'direct labels,' focusing on the structural representation of data rather than executable logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 20, 2026, 06:45 AM