Academic Figure Prompt
Pass
Audited by Gen Agent Trust Hub on Mar 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it is instructed to process untrusted data from user-provided documents to extract architectural concepts.\n
- Ingestion points: The workflow requires reading user-provided paper source files (LaTeX, Word, PDF) in Step 1 of SKILL.md.\n
- Boundary markers: The instructions do not specify any delimiters or safety warnings to help the agent distinguish between paper content and its own operational instructions.\n
- Capability inventory: The skill metadata includes the
bashtool, providing a potential capability surface if an injection attack occurs through processed files.\n - Sanitization: No sanitization or validation logic is defined for the content extracted from the external documents.\n- [SAFE]: The skill includes references to reputable well-known design and color services, including Adobe Color, ColorBrewer, and Coolors. These links are used for their intended primary purpose of assisting with academic figure design and represent no security risk.
Audit Metadata