html-slides

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use standard system openers such as 'open' on macOS, 'xdg-open' on Linux, and 'start' on Windows to provide users with a visual preview of their generated slides.
  • [EXTERNAL_DOWNLOADS]: Typography assets are retrieved from well-known and trusted services, specifically Google Fonts and Fontshare, which is a standard practice for web-based presentations.
  • [REMOTE_CODE_EXECUTION]: The skill recommends the use of the well-established 'python-pptx' library for extracting content from PowerPoint files during the conversion process.
  • [PROMPT_INJECTION]: The skill identifies a potential surface for indirect prompt injection due to its core functionality of processing external data.
  • Ingestion points: User-provided presentation text and uploaded .pptx files.
  • Boundary markers: Not explicitly defined within the conversion instructions.
  • Capability inventory: Includes writing local HTML files, executing platform-specific file openers, and running Python extraction scripts.
  • Sanitization: The skill does not explicitly detail sanitization procedures for the data extracted from PowerPoint files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 10:29 AM
Security Audit — agent-trust-hub — html-slides