excalidraw-diagram
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions require the agent to execute a local Python script (
render_excalidraw.py) and setup commands (playwright install chromium) using theuvpackage manager. These operations are essential for the skill's primary function of converting diagram JSON into viewable PNG images for quality validation. - [EXTERNAL_DOWNLOADS]: The skill performs external downloads during its setup phase (fetching the Chromium browser via Playwright) and at runtime (importing the Excalidraw library from the
esm.shCDN). These resources are sourced from well-known providers and are necessary for the rendering engine to function. - [DYNAMIC_EXECUTION]: The rendering script utilizes Playwright to automate a headless browser environment, executing JavaScript within a local HTML template to process diagram data and generate SVG/PNG exports.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and render
.excalidrawJSON files which may contain untrusted text. The agent is instructed to 'view' the rendered output to perform visual audits, creating an attack surface where instructions embedded in diagram text could influence agent behavior, though this is treated as a low-risk inherent characteristic of the task.
Audit Metadata