slide-generation
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
extract_paper_elements.pyto parse LaTeX source files and generate presentation skeletons. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface due to processing untrusted LaTeX input.
- Ingestion points: The
extract_paper_elements.pyscript reads and extracts text from user-provided.texfiles. - Boundary markers: None. The extracted content is placed directly into JSON or LaTeX output without delimiters to distinguish it from instructions.
- Capability inventory: The skill can read local files and execute Python scripts via the shell.
- Sanitization: The script does not sanitize or filter extracted text (e.g., titles, captions, or section content) for malicious agent instructions.
Audit Metadata