frontend-slides

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes platform-specific shell commands (open, xdg-open, start) to display generated HTML presentations and utilizes python3 to execute content extraction scripts.
  • [EXTERNAL_DOWNLOADS]: Recommends the installation of the python-pptx library and fetches typography assets from established services such as Google Fonts and Fontshare.
  • [PROMPT_INJECTION]: Ingests external data from PowerPoint files to generate HTML and JavaScript content, creating a surface for indirect prompt injection. (1) Ingestion points: user-supplied .ppt and .pptx files referenced in SKILL.md. (2) Boundary markers: No delimiters or ignore instructions are specified for extracted content. (3) Capability inventory: Performs subprocess calls via python and system openers, and writes generated code to the local file system. (4) Sanitization: No explicit filtering or validation of extracted text is described before interpolation into the final deck code.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 07:43 AM
Security Audit — agent-trust-hub — frontend-slides