skills/ilteoood/skills/drawio/Gen Agent Trust Hub

drawio

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The diagram export script (scripts/drawio-to-png.mjs) fetches the official draw.io rendering engine from https://viewer.diagrams.net/js/viewer-static.min.js to process diagrams when the local CLI is not available.
  • [COMMAND_EXECUTION]: The script attempts to locate and execute the drawio executable on the host system to perform exports by searching common installation paths and using system utilities.
  • [REMOTE_CODE_EXECUTION]: The skill launches a headless browser using Puppeteer to execute the official draw.io viewer's JavaScript code within a browser context for diagram rendering.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes diagram XML data.
  • Ingestion points: Reads contents from files with the .drawio extension in scripts/drawio-to-png.mjs.
  • Boundary markers: None identified; the XML content is processed directly for rendering.
  • Capability inventory: The rendering script has access to spawnSync for subprocess execution and writeFileSync for file system operations.
  • Sanitization: Content is escaped for JavaScript template literal embedding, but no semantic sanitization of the diagram XML structure is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 01:16 PM
Security Audit — agent-trust-hub — drawio