latex-paper-conversion
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill workflow involves executing shell commands, including pdflatex and bibtex, to compile documents and verify outputs. Processing untrusted LaTeX files via these tools can be risky if the compilation environment permits shell execution (e.g., write18).
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to dynamically create and run a Python script (convert_format.py) to perform content extraction and transformation. This represents dynamic script generation and execution at runtime, which is a functional part of the conversion process.
- [PROMPT_INJECTION]: The skill processes untrusted external LaTeX files, creating a surface for indirect prompt injection. 1. Ingestion points: Source .tex files provided by the user and mentioned in SKILL.md. 2. Boundary markers: The skill does not define specific boundaries or markers to isolate the LaTeX content from the agent's instructions. 3. Capability inventory: Python code execution, shell command execution (pdflatex), and file system write access. 4. Sanitization: The skill lacks instructions for sanitizing or validating the contents of the LaTeX files before parsing.
Audit Metadata