placeholder-images

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by incorporating external data into generated files.
  • Ingestion points: Processes user-provided labels and extracts design tokens from local CSS files (e.g., src/styles/main.css).
  • Boundary markers: Commands do not specify delimiters or instructions to ignore embedded content in labels or CSS.
  • Capability inventory: Utilizes Bash to run a Node.js script and Write to save the resulting SVG files to the filesystem.
  • Sanitization: Presence of sanitization for SVG tags or shell characters in labels is unverified.\n- [COMMAND_EXECUTION]: Executes a local Node.js utility script (scripts/quality/generate-placeholder.js) to perform image generation tasks. This is a vendor-internal script associated with the skill's primary functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 06:11 AM