lov-any2pdf

Warn

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: MEDIUMPRIVILEGE_ESCALATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The documentation in README.md and SKILL.md instructs the user or agent to run sudo apt install to install necessary font packages on Linux. This involves requesting administrative privileges for system-level modifications.\n- [EXTERNAL_DOWNLOADS]: The md2pdf.py script is designed to download remote assets at runtime. It automatically fetches Twemoji icons from Cloudflare's CDN and retrieves images from arbitrary URLs found within the input Markdown documents.\n- [INDIRECT_PROMPT_INJECTION]: As the skill processes untrusted Markdown data, it is susceptible to indirect prompt injection where malicious content in the document could influence the agent's behavior or trigger unintended network fetches via image tags.\n- [COMMAND_EXECUTION]: The skill's workflow depends on the agent executing shell commands for setting up the environment (using pip and npx) and generating the PDF (using the Python script or a pandoc fallback).
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 30, 2026, 03:43 AM
Security Audit — agent-trust-hub — lov-any2pdf