slides-grab-design

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the slides-grab CLI and yt-dlp to manage slide lifecycle tasks such as validation (slides-grab validate), asset fetching (slides-grab fetch-video), and launching the editor (slides-grab edit). It also invokes node scripts/build-viewer.js as part of its workflow.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of images and videos from external web sources to the local project directory using the slides-grab image and slides-grab fetch-video commands. It also references web-hosted libraries (Pretendard, Chart.js, Mermaid) via common CDNs for slide rendering.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes user-provided outlines and edits to generate executable HTML and CSS code.
  • Ingestion points: Processes slide-outline.md, requested text edits, and external video URLs.
  • Boundary markers: No specific delimiters or safety warnings for embedded instructions are defined in the instructions.
  • Capability inventory: Includes file system writing, shell command execution via the slides-grab tool, and network requests for downloading media.
  • Sanitization: The provided instructions do not specify any validation or sanitization routines for the ingested data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 12:01 PM
Security Audit — agent-trust-hub — slides-grab-design