drawio
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes instructions for using the
drawio-exportanddrawiocommand-line tools to convert diagram files into PNG, SVG, or PDF formats. - [INDIRECT_PROMPT_INJECTION]: The skill processes XML-based diagram files which could potentially contain malicious instructions embedded within text labels.
- Ingestion points: Untrusted content enters the agent context through the
valueattribute ofmxCellelements within.drawiofiles. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content within these labels.
- Capability inventory: The skill possesses the capability to execute shell commands (
drawio-export) as defined inSKILL.md. - Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the XML content before processing.
Audit Metadata