html-presentations

Pass

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches front-end styling and diagramming libraries (Tailwind CSS, Mermaid.js) and typography from well-known public CDNs (jsDelivr, Google Fonts).\n- [COMMAND_EXECUTION]: Executes a local bash script (create_standalone.sh) to automate the embedding of image files into the presentation as Base64 data URIs. It also invokes the open command to display the generated HTML file to the user.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided markdown specification files to generate presentation content.\n
  • Ingestion points: Reads the markdown spec file (e.g., spec.md).\n
  • Boundary markers: Uses specific slide delimiters (* * *) and standard markdown syntax to structure the output.\n
  • Capability inventory: Includes file system write access for the resulting HTML and shell execution for the build process.\n
  • Sanitization: The agent converts markdown structural elements into a predefined set of HTML tags and CSS classes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 30, 2026, 02:50 AM
Security Audit — agent-trust-hub — html-presentations