presentation-builder

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a secure generation pipeline that automatically sanitizes user-provided HTML content. It removes potentially dangerous elements such as <script>, <iframe>, <object>, and all inline event handlers (e.g., onclick) from the input blocks before generating the final presentation.
  • [SAFE]: The generated browser runtime includes a dedicated sanitization function (sanitizeEditableHtml) and a URL safety predicate (isSafeHref). These ensure that even content modified via 'Edit mode' in the browser is cleaned of unauthorized tags and that links are restricted to safe protocols, specifically blocking the javascript: scheme and obfuscated path traversals.
  • [SAFE]: Data persistence is handled securely using the browser's localStorage API under a specific, versioned namespace. This keeps presentation drafts local to the user's environment and prevents data leakage to external services.
  • [SAFE]: The utility scripts (create-presentation.cjs and validate-presentation.cjs) are dependency-free and rely only on standard Node.js modules for file I/O and process management. They do not perform any unauthorized network requests or access sensitive system credentials.
  • [SAFE]: The testing suite utilizes the Node.js vm module to perform static analysis and unit testing on the generated runtime code. This allows the author to verify that the security logic (like the XSS sanitizer) works correctly without actually executing potentially untrusted code in a full browser DOM.
  • [SAFE]: The skill follows a strict 'self-contained' policy by default, rejecting external network assets (fonts, images, scripts) unless the user explicitly opts in. This minimizes the risk of supply chain attacks and ensures the presentations remain functional in offline environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 08:16 PM