skills/minimax-ai/skills/minimax-pdf/Gen Agent Trust Hub

minimax-pdf

Pass

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Several scripts (fill_inspect.py, fill_write.py, merge.py, and render_body.py) include an automated dependency check that installs the pypdf library via pip if it is not present. The make.sh utility also automates the installation of reportlab, matplotlib, and the playwright framework. These are all established, well-known software components.\n- [COMMAND_EXECUTION]: The skill orchestrates multiple Python and Node.js sub-processes through its make.sh script. This includes the use of a Python heredoc to dynamically modify configuration files and the execution of the Playwright browser engine for cover rendering.\n- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface within its document parsing features. When using the reformat functionality, the agent extracts text from external documents (PDF, Markdown, or text) which may contain instructions intended to influence the agent's behavior.\n
  • Ingestion points: reformat_parse.py (via parse_markdown and parse_pdf).\n
  • Boundary markers: None; the agent processes the extracted content without specific delimiters to ignore embedded instructions.\n
  • Capability inventory: Subprocess execution, automated library installation, and file system writes (PDF generation).\n
  • Sanitization: None; the parsing logic focuses on structure and formatting but does not filter for instructional content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 22, 2026, 07:15 AM