verified-infographic-generator
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill architecture prioritizes factual accuracy by enforcing a workflow that separates data verification from visual rendering. This design prevents the autonomous creation of misleading or fabricated content.- [PROMPT_INJECTION]: The workflow involves ingesting data via web searches in Phase 2 (Source Discovery). This potential surface for indirect prompt injection is addressed through rigorous cross-referencing requirements against peer-reviewed journals and institutional reports, followed by a mandatory user-validation checkpoint in Phase 4.- [COMMAND_EXECUTION]: The skill performs legitimate file operations to save audit trails and artifacts in a project-specific directory (docs/posters/). No unauthorized or dangerous command execution patterns, such as privilege escalation or persistence mechanisms, were identified.- [EXTERNAL_DOWNLOADS]: The skill references well-known and trusted text-to-image services (OpenAI, Gemini, DALL·E) for final rendering, which is considered safe behavior under standard operation.
Audit Metadata