presenting-conference-talks
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted research papers (PDF or LaTeX source) as data to generate presentation content. \n
- Ingestion points: Research paper files processed during the content extraction phase in SKILL.md. \n
- Boundary markers: No explicit delimiters or instructions to ignore embedded instructions within the source paper are defined in the templates. \n
- Capability inventory: The skill writes LaTeX and Python files to the filesystem and executes build commands (latexmk, python3) as part of its slide generation workflow. \n
- Sanitization: Paper content is interpolated into slide templates without explicit sanitization or filtering routines.
- [COMMAND_EXECUTION]: The skill workflow relies on the execution of local build commands for document compilation. \n
- Evidence: Usage instructions for
latexmk -pdf slides.texandpython3 generate_slides.pyare provided for generating the final Beamer and PPTX outputs. - [EXTERNAL_DOWNLOADS]: The skill specifies a dependency on the
python-pptxlibrary, which is a standard and well-known package available through the official Python Package Index (PyPI).
Audit Metadata