academic-latex

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill executes local binaries including pdflatex, xelatex, and lualatex. These engines are Turing-complete and can interact with the file system or execute scripts (in the case of lualatex).
  • [PROMPT_INJECTION] (LOW): Vulnerable to Indirect Prompt Injection (Category 8) as it processes untrusted user-supplied document content. * Ingestion points: User-provided LaTeX content via the /compile endpoint or .tex files. * Boundary markers: Absent; there are no instructions to ignore embedded agent commands within the LaTeX source. * Capability inventory: Subprocess execution of LaTeX engines, file system writes to /workspace/output/, and local network requests via curl. * Sanitization: Absent; the skill relies on the default configuration of the TeX Live installation.
  • [EXTERNAL_DOWNLOADS] (SAFE): Network usage is limited to localhost:8080, which is a whitelisted destination.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 07:35 AM