Skills
Navigate
skills/vodailocz/kilo-kit-mcp/doc/Gen Agent Trust Hub

doc

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes system tools to perform document rendering.
  • In scripts/render_docx.py, the run_cmd_no_check function calls soffice and pdftoppm using argument lists.
  • SKILL.md provides documentation for similar shell commands intended for visual verification.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external .docx files, which is a potential surface for indirect prompt injection.
  • Ingestion points: scripts/render_docx.py (in functions calc_dpi_via_ooxml_docx and rasterize) and the recommended python-docx library.
  • Boundary markers: No specific delimiters or 'ignore instructions' markers are used when processing document content.
  • Capability inventory: The skill uses subprocess.run to call soffice and pdftoppm in scripts/render_docx.py (via run_cmd_no_check), and defines output paths in SKILL.md (tmp/docs/, output/doc/).
  • Sanitization: Command execution is performed via structured argument lists in scripts/render_docx.py, which prevents shell-level command injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 02:34 PM