anydesign
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on several local Python utility scripts located in the
scripts/directory. These includecapture_site.pyfor headless browser interaction,extract_css_vars.pyfor token harvesting, andexport_for_claude_design.pyfor report generation. These are invoked by the agent to perform deterministic analysis on input materials. - [EXTERNAL_DOWNLOADS]: The skill performs network requests to retrieve HTML and CSS from user-specified URLs. This functionality is essential for its primary purpose of website design analysis and does not involve downloading or executing untrusted code into the agent's main execution context.
- [SAFE]: Analysis of the source code for the provided Python scripts shows standard usage of libraries like Playwright, Pillow, and urllib. Secret management instructions (e.g., for Figma) follow safe practices by leveraging existing platform tools (Figma MCP) rather than requesting hardcoded credentials.
Audit Metadata