cro
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill's logic and instructions are entirely focused on its stated purpose of analyzing marketing pages for conversion improvements. No obfuscation, unauthorized exfiltration, or persistence mechanisms were found.
- [COMMAND_EXECUTION]: The skill provides a Node.js script in 'references/playwright-analysis.md' and instructs the agent to execute it via the command line to perform website audits. This is a functional requirement for analyzing live URLs.
- [EXTERNAL_DOWNLOADS]: The analysis setup involves installing the 'playwright' package and Chromium browser via NPM. These are official tools from Microsoft, which is a recognized and trusted service provider.
- [SAFE]: While the skill processes untrusted content from external websites (HTML and visible text), the instructions strictly scope this data to marketing analysis, presenting only a standard functional risk of indirect prompt injection which is typical for browsing-enabled agents.
Audit Metadata