academic-beamer-deck
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content including .tex files, figures, and presentation context to generate formatted slide decks.
- Ingestion points: Processes external .tex files, figure assets, and textual talk context as described in the frontmatter and workflow.
- Capability inventory: Generates output .tex files and instructs the agent to perform compilation using pdfLaTeX.
- Boundary markers: The instructions do not define specific delimiters or ignore instructions for the processed content.
- Sanitization: No explicit sanitization or validation steps for the input LaTeX code are specified.
- [DYNAMIC_EXECUTION]: The skill instructions specify runtime compilation of generated LaTeX source code.
- Evidence: The Verification section requires compilation with pdfLaTeX and multiple passes to resolve references. This is a standard and expected part of the document generation workflow for LaTeX.
Audit Metadata