drawio-reconstruction

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes subprocess.run in scripts/batch_verify.py and scripts/export_drawio.py to execute localized scripts and the Draw.io desktop CLI. These operations are used for validating file structure and generating image previews, which are core functions of the skill.
  • [EXTERNAL_DOWNLOADS]: The skill requires the Pillow library, which it instructs the agent to install from the official Python Package Index (PyPI). This is a well-known and trusted library for image processing.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided images and Draw.io XML files as input for reconstruction.
  • Ingestion points: Reference images are analyzed visually by the agent, and .drawio XML files are parsed using the scripts/check_drawio.py script.
  • Boundary markers: The skill mandates the creation of a reference inventory and audit files (<stem>.audit.md) to track elements, which helps the agent distinguish between data observations and instructional goals.
  • Capability inventory: The skill has the capability to write files and execute local subprocesses to verify diagram integrity and export visuals.
  • Sanitization: scripts/check_drawio.py uses the xml.etree.ElementTree parser to validate the structural integrity of the XML files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 01:25 PM
Security Audit — agent-trust-hub — drawio-reconstruction