design-lookup

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a Python script (scripts/fetch-page.py) that leverages the Playwright library to automate a browser for web scraping. While this involves executing local scripts and potentially interacting with web content, the implementation is focused on specific design-related selectors and uses standard command-line arguments.
  • [EXTERNAL_DOWNLOADS]: The skill fetches data from external websites (CodePen, UIVerse) and retrieves raw SVG files from well-known GitHub repositories (Lucide, Heroicons, Tabler, Feather, Simple Icons). These downloads are used to provide the user with design assets and are directed at established, trusted developer platforms.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the web (HTML/CSS from scraped pages). While this presents a surface for indirect prompt injection, the impact is low as the agent is instructed to clean the code and present it to the user as a snippet rather than executing it. The boundary is maintained by the agent's task of 'cleaning up' and 'presenting results'.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 06:16 AM
Security Audit — agent-trust-hub — design-lookup