redis-excalidraw-diagrams

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the Playwright library and the Chromium browser to facilitate diagram rendering. These are standard tools for browser automation maintained by Microsoft.
  • [EXTERNAL_DOWNLOADS]: The rendering template fetches the Excalidraw library from esm.sh, which is a well-known CDN for JavaScript modules.
  • [COMMAND_EXECUTION]: The skill utilizes Python and Playwright to automate a headless browser for rendering diagrams from JSON data. This behavior is transparently described in the documentation and is necessary for the skill's primary function.
  • [PROMPT_INJECTION]: The skill processes user architectural requirements to generate Excalidraw JSON, which is then rendered in a browser context. While this presents an indirect prompt injection surface where malicious instructions could theoretically be embedded in the processed data, the impact is confined to the headless browser environment and is a known trade-off for this type of rendering functionality.
  • Ingestion points: User requests for diagrams are used to generate content for SKILL.md workflows and the Excalidraw JSON.
  • Boundary markers: The skill provides structured JSON templates in references/element-templates.md and a schema in references/json-schema.md to guide generation.
  • Capability inventory: references/render_excalidraw.py reads local files and executes a headless browser via Playwright.
  • Sanitization: No explicit sanitization of diagram text is performed in the provided scripts; it relies on the safety properties of the Excalidraw library and the JSON format.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 06:25 PM