case-classroom-test

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements tools for business case analysis and reporting, with implementation logic that is consistent with its stated objectives.
  • [COMMAND_EXECUTION]: The skill's rendering script, render-reader-formats.ts, spawns subprocesses to execute local browsers (Chrome/Chromium) and system file openers. These actions are properly scoped to the purpose of converting and opening generated documents.
  • [EXTERNAL_DOWNLOADS]: The skill references the Mermaid.js library from the JSDelivr CDN (cdn.jsdelivr.net), a well-known service, to enable diagram rendering in a headless environment. This download is used exclusively for document formatting.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided business cases and teaching notes as defined in SKILL.md. It includes security measures like safeUrl and escapeHtml in its rendering logic to prevent malicious content in the input files from affecting the generated output copies.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 02:43 AM