drawio

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to construct and execute shell commands using user-provided diagram names and export formats without explicit guidance on sanitization or escaping.
  • Evidence: The skill uses user-supplied placeholders <name> and <format> directly in shell command templates: drawio -x -f <format> -e -b 10 -o "<DRAWIO_OUTPUT_DIR>/<name>/<name>.drawio.<format>" <input.drawio>.
  • Risk: If a user provides a diagram name containing shell metacharacters (e.g., diagram; rm -rf /), the agent might execute unintended commands depending on the shell environment and how the agent handles the Bash tool calls.
  • File Opening Risk: Similar risks exist for platform-specific file-opening commands (open <file>, xdg-open <file>, start <file>) where the file path is partially derived from user input.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 03:16 AM
Security Audit — agent-trust-hub — drawio