architecture-diagram

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's HTML template references the html2canvas and jsPDF libraries from the jsDelivr CDN. These references use Subresource Integrity (SRI) hashes to verify the integrity of the downloaded scripts when the generated file is opened in a browser.
  • [COMMAND_EXECUTION]: The skill workflow includes running internal Python scripts (geometry-audit.py, audit-labels.py, and coordinate-shifter.py) to validate the generated SVG geometry and labels. These scripts are part of the skill's resource directory and are executed on the locally generated diagram file.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user descriptions of system architectures to produce diagrams. The output is constrained to a static HTML/SVG format, and the agent uses its own auditing scripts to verify the output, providing a layer of validation for the generated content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 06:26 PM