skills/tjxj/z-skills/z-md-to-pdf/Gen Agent Trust Hub

z-md-to-pdf

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The scripts/setup.sh script fetches the TinyTeX distribution from the official GitHub releases page of the RStudio TinyTeX project (github.com/rstudio/tinytex-releases). This is a well-known and established service for LaTeX environments.
  • [COMMAND_EXECUTION]: The skill uses shell scripts to automate environment preparation and document building. It invokes system tools like pandoc, xelatex, tlmgr, and package managers (apt-get, brew). These operations are consistent with the skill's primary purpose of document transformation and environment setup.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted Markdown files and LaTeX projects provided by the user for conversion.
  • Ingestion points: Markdown input files processed in scripts/build.sh and external LaTeX project directories in scripts/build-upstream.sh.
  • Boundary markers: None explicitly defined for the processing phase.
  • Capability inventory: The skill has file-system read/write access and executes shell commands via Pandoc and XeLaTeX.
  • Sanitization: Basic sanitization is performed on Markdown image syntax, but the skill relies on the security of the underlying Pandoc and XeLaTeX engines to handle potentially adversarial content within the input files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 11:05 PM
Security Audit — agent-trust-hub — z-md-to-pdf