skills/pengelbrecht/skills/gws-slides/Gen Agent Trust Hub

gws-slides

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes multiple shell commands using the 'gws' CLI for Google Slides and Drive operations, including a command to share Drive files with 'anyone' to allow the Slides API to access images. It also utilizes 'curl' and image processing tools like 'rsvg-convert' and 'ImageMagick'.
  • [EXTERNAL_DOWNLOADS]: Fetches slide thumbnail images from Google-hosted URLs using 'curl' for visual verification during the quality check phase.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it lacks safeguards for handling user-supplied text used in slide content.
  • Ingestion points: User prompts (e.g., 'presentation about X') and data retrieved from existing presentations.
  • Boundary markers: Absent; user content is directly interpolated into API payloads.
  • Capability inventory: 'gws' CLI allows for full creation, retrieval, and modification of Google Slides and Drive files.
  • Sanitization: No input validation or escaping is implemented for user-provided strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 01:44 PM
Security Audit — agent-trust-hub — gws-slides